Références:
abstract_narrative_understanding
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/abstract_narrative_understanding')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 3000 |
'train' | 2400 |
'validation' | 600 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
anachronismes
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/anachronisms')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 230 |
'train' | 184 |
'validation' | 46 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
similarité_analogique
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/analogical_similarity')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 323 |
'train' | 259 |
'validation' | 64 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
implication_analytique
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/analytic_entailment')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 70 |
'train' | 54 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
arithmétique
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/arithmetic')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 15023 |
'train' | 12019 |
'validation' | 3004 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
ascii_word_recognition
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/ascii_word_recognition')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 5000 |
'train' | 4000 |
'validation' | 1000 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
authorship_verification
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/authorship_verification')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 880 |
'train' | 704 |
'validation' | 176 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
catégorisation_auto
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/auto_categorization')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 328 |
'train' | 263 |
'validation' | 65 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
auto_debugging
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/auto_debugging')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 34 |
'train' | 18 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
barbecue_lite_json
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/bbq_lite_json')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 16076 |
'train' | 12866 |
'validation' | 3210 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
pontage_anaphore_résolution_barqa
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/bridging_anaphora_resolution_barqa')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 648 |
'train' | 519 |
'validation' | 129 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
jugement_causal
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/causal_judgment')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 190 |
'train' | 152 |
'validation' | 38 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
cause et effet
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/cause_and_effect')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 153 |
'train' | 123 |
'validation' | 30 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
échec et mat_en_un
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/checkmate_in_one')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 3498 |
'train' | 2799 |
'validation' | 699 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
chess_state_tracking
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/chess_state_tracking')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 6000 |
'train' | 4800 |
'validation' | 1200 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
théorème_du_reste_chinois
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/chinese_remainder_theorem')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 500 |
'train' | 400 |
'validation' | 100 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
cifar10_classification
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/cifar10_classification')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 20000 |
'train' | 16000 |
'validation' | 4000 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
code_line_description
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/code_line_description')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 60 |
'train' | 44 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
noms de code
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/codenames')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 85 |
'train' | 68 |
'validation' | 17 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
Couleur
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/color')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 4000 |
'train' | 3200 |
'validation' | 800 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
morphème_commun
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/common_morpheme')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 50 |
'train' | 34 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
combinaisons_conceptuelles
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/conceptual_combinations')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 103 |
'train' | 84 |
'validation' | 19 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
conlang_translation
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/conlang_translation')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 164 |
'train' | 132 |
'validation' | 32 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
contexte_parametric_knowledge_conflicts
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/contextual_parametric_knowledge_conflicts')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 17528 |
'train' | 14023 |
'validation' | 3505 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
crash_blossom
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/crash_blossom')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 38 |
'train' | 22 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
crass_ai
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/crass_ai')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 44 |
'train' | 28 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
cryobiologie_espagnol
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/cryobiology_spanish')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 146 |
'train' | 117 |
'validation' | 29 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
cryptonite
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/cryptonite')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 26157 |
'train' | 20926 |
'validation' | 5231 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
cs_algorithms
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/cs_algorithms')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1320 |
'train' | 1056 |
'validation' | 264 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
dark_humor_detection
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/dark_humor_detection')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 80 |
'train' | 64 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
date_compréhension
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/date_understanding')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 369 |
'train' | 296 |
'validation' | 73 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
désambiguïsation_qa
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/disambiguation_qa')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 258 |
'train' | 207 |
'validation' | 51 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
discours_marker_prediction
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/discourse_marker_prediction')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 857 |
'train' | 686 |
'validation' | 171 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
disfl_qa
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/disfl_qa')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 8000 |
'train' | 6400 |
'validation' | 1600 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
dyck_langues
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/dyck_languages')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1000 |
'train' | 800 |
'validation' | 200 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
mathématiques_élémentaires_qa
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/elementary_math_qa')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 38160 |
'train' | 30531 |
'validation' | 7629 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
emoji_movie
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/emoji_movie')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 100 |
'train' | 80 |
'validation' | 20 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
emojis_emotion_prediction
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/emojis_emotion_prediction')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 131 |
'train' | 105 |
'validation' | 26 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
jugements_empiriques
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/empirical_judgments')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 99 |
'train' | 80 |
'validation' | 19 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
proverbes_anglais
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/english_proverbs')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 34 |
'train' | 18 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
proverbes_anglais_russes
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/english_russian_proverbs')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 80 |
'train' | 64 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
polarité_entraînée
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/entailed_polarity')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 148 |
'train' | 119 |
'validation' | 29 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
entraîné_polarité_hindi
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/entailed_polarity_hindi')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 138 |
'train' | 111 |
'validation' | 27 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
raisonnement_épistémique
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/epistemic_reasoning')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 2000 |
'train' | 1600 |
'validation' | 400 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
évaluation_de_l'essentiel_de_l'information
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/evaluating_information_essentiality')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 68 |
'train' | 52 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
fact_checker
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/fact_checker')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 7154 |
'train' | 5724 |
'validation' | 1430 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
raisonnement_fantastique
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/fantasy_reasoning')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 201 |
'train' | 161 |
'validation' | 40 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
quelques_shot_nlg
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/few_shot_nlg')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 153 |
'train' | 123 |
'validation' | 30 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
figure_of_speech_detection
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/figure_of_speech_detection')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 59 |
'train' | 43 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
formal_fallacies_syllogisms_negation
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/formal_fallacies_syllogisms_negation')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 14200 |
'train' | 11360 |
'validation' | 2840 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
gemme
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/gem')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 14802 |
'train' | 11845 |
'validation' | 2957 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
gender_inclusive_sentences_allemand
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/gender_inclusive_sentences_german')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 200 |
'train' | 160 |
'validation' | 40 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
culture générale
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/general_knowledge')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 70 |
'train' | 54 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
formes géométriques
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/geometric_shapes')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 359 |
'train' | 288 |
'validation' | 71 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
objectif_étape_wikihow
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/goal_step_wikihow')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 7053 |
'train' | 5643 |
'validation' | 1410 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
gre_reading_comprehension
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/gre_reading_comprehension')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 31 |
'train' | 15 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
hhh_alignement
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/hhh_alignment')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 221 |
'train' | 179 |
'validation' | 42 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
hindi_question_answering
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/hindi_question_answering')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 6610 |
'train' | 5288 |
'validation' | 1322 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
connaissance_hindoue
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/hindu_knowledge')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 175 |
'train' | 140 |
'validation' | 35 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
hinglish_toxicity
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/hinglish_toxicity')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 200 |
'train' | 160 |
'validation' | 40 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
organes_humains_sens
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/human_organs_senses')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 42 |
'train' | 26 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
hyperbate
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/hyperbaton')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 50000 |
'train' | 40000 |
'validation' | 10000 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
identifier_math_theorems
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/identify_math_theorems')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 53 |
'train' | 37 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
identifier_odd_metaphor
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/identify_odd_metaphor')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 47 |
'train' | 31 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
implicatures
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/implicatures')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 492 |
'train' | 394 |
'validation' | 98 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
relations_implicites
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/implicit_relations')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 85 |
'train' | 68 |
'validation' | 17 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
reconnaissance_intention
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/intent_recognition')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 693 |
'train' | 555 |
'validation' | 138 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
international_phonetic_alphabet_nli
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/international_phonetic_alphabet_nli')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 126 |
'train' | 101 |
'validation' | 25 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
international_phonetic_alphabet_transliterate
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/international_phonetic_alphabet_transliterate')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1003 |
'train' | 803 |
'validation' | 200 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
intersect_geometry
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/intersect_geometry')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 249999 |
'train' | 200000 |
'validation' | 49999 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
ironie_identification
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/irony_identification')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 99 |
'train' | 80 |
'validation' | 19 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
kanji_ascii
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/kanji_ascii')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1092 |
'train' | 875 |
'validation' | 217 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
kannada
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/kannada')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 316 |
'train' | 253 |
'validation' | 63 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
key_value_maps
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/key_value_maps')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 101 |
'train' | 80 |
'validation' | 21 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
inconnus_connus
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/known_unknowns')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 46 |
'train' | 30 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
language_games
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/language_games')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 2128 |
'train' | 1704 |
'validation' | 424 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
identification_langue
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/language_identification')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 10000 |
'train' | 8000 |
'validation' | 2000 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
mappings_linguistiques
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/linguistic_mappings')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 15527 |
'train' | 12426 |
'validation' | 3101 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
puzzles_linguistiques
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/linguistics_puzzles')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 2000 |
'train' | 1600 |
'validation' | 400 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
list_functions
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/list_functions')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 10750 |
'train' | 8700 |
'validation' | 2050 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
grille_logique_puzzle
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/logic_grid_puzzle')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1000 |
'train' | 800 |
'validation' | 200 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
logical_args
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/logical_args')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 32 |
'train' | 16 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
déduction logique
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/logical_deduction')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1500 |
'train' | 1200 |
'validation' | 300 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
logical_fallacy_detection
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/logical_fallacy_detection')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 2800 |
'train' | 2240 |
'validation' | 560 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
suite logique
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/logical_sequence')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 39 |
'train' | 23 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
Induction mathematique
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/mathematical_induction')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 69 |
'train' | 53 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
formes de matrice
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/matrixshapes')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 4462 |
'train' | 3570 |
'validation' | 892 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
metaphor_boolean
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/metaphor_boolean')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 680 |
'train' | 544 |
'validation' | 136 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
métaphore_compréhension
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/metaphor_understanding')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 234 |
'train' | 188 |
'validation' | 46 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
minute_mysteries_qa
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/minute_mysteries_qa')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 477 |
'train' | 383 |
'validation' | 94 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
idées fausses
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/misconceptions')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 219 |
'train' | 176 |
'validation' | 43 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
idées fausses_russe
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/misconceptions_russian')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 49 |
'train' | 33 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
mnist_ascii
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/mnist_ascii')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 69984 |
'train' | 55988 |
'validation' | 13996 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
modification_arithmétique
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/modified_arithmetic')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 6000 |
'train' | 4800 |
'validation' | 1200 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
permission_morale
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/moral_permissibility')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 342 |
'train' | 274 |
'validation' | 68 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
movie_dialog_same_or_different
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/movie_dialog_same_or_different')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 50000 |
'train' | 40000 |
'validation' | 10000 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
movie_recommendation
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/movie_recommendation')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 500 |
'train' | 400 |
'validation' | 100 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
multi_data_wrangling
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/mult_data_wrangling')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 7854 |
'train' | 6380 |
'validation' | 1474 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
multiémo
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/multiemo')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1437281 |
'train' | 1149873 |
'validation' | 287408 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
instructions_naturelles
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/natural_instructions')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 193250 |
'train' | 154615 |
'validation' | 38635 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
naviguer
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/navigate')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1000 |
'train' | 800 |
'validation' | 200 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
mots_non-sens_grammaire
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/nonsense_words_grammar')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 50 |
'train' | 34 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
roman_concepts
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/novel_concepts')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 32 |
'train' | 16 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
object_counting
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/object_counting')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1000 |
'train' | 800 |
'validation' | 200 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
odd_one_out
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/odd_one_out')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 86 |
'train' | 69 |
'validation' | 17 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
les opérateurs
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/operators')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 210 |
'train' | 168 |
'validation' | 42 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
paragraphe_segmentation
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/paragraph_segmentation')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 9000 |
'train' | 7200 |
'validation' | 1800 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
parsinlu_qa
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/parsinlu_qa')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1050 |
'train' | 840 |
'validation' | 210 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
parsinlu_reading_comprehension
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/parsinlu_reading_comprehension')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 518 |
'train' | 415 |
'validation' | 103 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
pingouins_dans_une_table
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/penguins_in_a_table')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 149 |
'train' | 120 |
'validation' | 29 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
éléments_périodiques
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/periodic_elements')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 654 |
'train' | 524 |
'validation' | 130 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
idiomes_persans
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/persian_idioms')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 66 |
'train' | 50 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
expression_relative
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/phrase_relatedness')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 100 |
'train' | 80 |
'validation' | 20 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
intuition_physique
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/physical_intuition')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 81 |
'train' | 65 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
la physique
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/physics')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 229 |
'train' | 184 |
'validation' | 45 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
questions_physiques
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/physics_questions')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 54 |
'train' | 38 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
play_dialog_same_or_different
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/play_dialog_same_or_different')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 3264 |
'train' | 2612 |
'validation' | 652 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
étiquetage_de_la_séquence_polonaise
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/polish_sequence_labeling')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 12812 |
'train' | 10250 |
'validation' | 2562 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
présuppositions_as_nli
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/presuppositions_as_nli')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 735 |
'train' | 588 |
'validation' | 147 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
qa_wikidata
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/qa_wikidata')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 20321 |
'train' | 16257 |
'validation' | 4064 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
question_selection
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/question_selection')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1582 |
'train' | 1266 |
'validation' | 316 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
vrai_ou_faux_texte
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/real_or_fake_text')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 15088 |
'train' | 12072 |
'validation' | 3016 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
raisonnement_about_colored_objects
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/reasoning_about_colored_objects')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 2000 |
'train' | 1600 |
'validation' | 400 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
repeat_copy_logic
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/repeat_copy_logic')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 32 |
'train' | 16 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
reformuler
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/rephrase')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 78 |
'train' | 62 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
devinette_sens
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/riddle_sense')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 49 |
'train' | 33 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
ruine_names
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/ruin_names')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 448 |
'train' | 359 |
'validation' | 89 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
saillant_translation_error_detection
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/salient_translation_error_detection')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 998 |
'train' | 799 |
'validation' | 199 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
communiqué_de_presse_scientifique
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/scientific_press_release')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 50 |
'train' | 34 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
semantic_parsing_in_context_sparc
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/semantic_parsing_in_context_sparc')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1155 |
'train' | 924 |
'validation' | 231 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
semantic_parsing_spider
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/semantic_parsing_spider')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1034 |
'train' | 828 |
'validation' | 206 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
phrase_ambiguïté
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/sentence_ambiguity')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 60 |
'train' | 44 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
similitudes_abstraction
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/similarities_abstraction')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 76 |
'train' | 60 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
simp_turing_concept
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/simp_turing_concept')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 6390 |
'train' | 5112 |
'validation' | 1278 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
simple_arithmetic_json
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/simple_arithmetic_json')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 30 |
'train' | 14 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
simple_arithmetic_json_multiple_choice
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/simple_arithmetic_json_multiple_choice')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 8 |
'train' | 0 |
'validation' | 0 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
simple_arithmetic_json_subtasks
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/simple_arithmetic_json_subtasks')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 30 |
'train' | 15 |
'validation' | 15 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
simple_arithmetic_multiple_targets_json
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/simple_arithmetic_multiple_targets_json')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | dix |
'train' | 0 |
'validation' | 0 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
questions_éthiques_simples
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/simple_ethical_questions')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 115 |
'train' | 92 |
'validation' | 23 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
simple_text_editing
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/simple_text_editing')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 47 |
'train' | 31 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
grognements
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/snarks')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 181 |
'train' | 145 |
'validation' | 36 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
social_iqa
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/social_iqa')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1935 |
'train' | 1548 |
'validation' | 387 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
aide sociale
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/social_support')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 897 |
'train' | 718 |
'validation' | 179 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
compréhension_sportive
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/sports_understanding')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 986 |
'train' | 789 |
'validation' | 197 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
histoires_étranges
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/strange_stories')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 174 |
'train' | 140 |
'validation' | 34 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
stratégieqa
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/strategyqa')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 2289 |
'train' | 1832 |
'validation' | 457 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
informations_suffisantes
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/sufficient_information')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 39 |
'train' | 23 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
suicide_risk
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/suicide_risk')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 40 |
'train' | 24 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
swahili_anglais_proverbes
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/swahili_english_proverbs')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 153 |
'train' | 123 |
'validation' | 30 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
proverbes_suédois_vers_allemands
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/swedish_to_german_proverbs')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 72 |
'train' | 56 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
symbole_interprétation
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/symbol_interpretation')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 990 |
'train' | 795 |
'validation' | 195 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
séquences_temporelles
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/temporal_sequences')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1000 |
'train' | 800 |
'validation' | 200 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
tendu
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/tense')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 286 |
'train' | 229 |
'validation' | 57 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
cadran horaire
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/timedial')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 2550 |
'train' | 2040 |
'validation' | 510 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
topic_chat
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/topical_chat')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 22295 |
'train' | 17836 |
'validation' | 4459 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
tracking_shuffled_objects
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/tracking_shuffled_objects')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 3750 |
'train' | 3000 |
'validation' | 750 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
compréhension_fables
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/understanding_fables')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 189 |
'train' | 152 |
'validation' | 37 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
annuler_permutation
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/undo_permutation')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 300 |
'train' | 240 |
'validation' | 60 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
conversion d'unité
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/unit_conversion')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 23936 |
'train' | 19151 |
'validation' | 4785 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
unité_interprétation
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/unit_interpretation')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 100 |
'train' | 80 |
'validation' | 20 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
unnatural_in_context_learning
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/unnatural_in_context_learning')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 73420 |
'train' | 58736 |
'validation' | 14684 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
vitaminc_fact_verification
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/vitaminc_fact_verification')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 54668 |
'train' | 43735 |
'validation' | 10933 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
qu'est-ce_que_le_tao
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/what_is_the_tao')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 36 |
'train' | 20 |
'validation' | 16 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
quel_wiki_edit
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/which_wiki_edit')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 571 |
'train' | 457 |
'validation' | 114 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
pourquoi
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/winowhy')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 2862 |
'train' | 2290 |
'validation' | 572 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
tri_de_mots
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/word_sorting')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 1900 |
'train' | 1520 |
'validation' | 380 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}
word_unscrambling
Utilisez la commande suivante pour charger cet ensemble de données dans TFDS :
ds = tfds.load('huggingface:bigbench/word_unscrambling')
- Descriptif :
The Beyond the Imitation Game Benchmark (BIG-bench) is a collaborative benchmark intended to
probe large language models, and extrapolate their future capabilities.
- Licence : Licence Apache 2.0
- Version : 0.0.0
- Fractionnements :
| Diviser | Exemples |
|---|---|
'default' | 8917 |
'train' | 7134 |
'validation' | 1783 |
- Caractéristiques :
{
"idx": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"inputs": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_targets": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"multiple_choice_scores": {
"feature": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}