|
{ |
|
"architectures": [ |
|
"ILKTModel" |
|
], |
|
"auto_map": { |
|
"AutoConfig": "config.ILKTConfig", |
|
"AutoModel": "model.ILKTModel" |
|
}, |
|
"backbone_config": { |
|
"pretrained_model_name_or_path": "google-bert/bert-base-multilingual-cased", |
|
"torch_dtype": "bfloat16", |
|
"trust_remote_code": true |
|
}, |
|
"cls_head_config": { |
|
"dropout": 0.0, |
|
"n_dense": 1, |
|
"pool_type": "cls", |
|
"use_batch_norm": true, |
|
"use_layer_norm": false |
|
}, |
|
"cls_heads": [ |
|
[ |
|
3, |
|
"allegro--klej-cdsc-e" |
|
], |
|
[ |
|
2, |
|
"allegro--klej-psc" |
|
], |
|
[ |
|
2, |
|
"allegro--klej-dyk" |
|
], |
|
[ |
|
5, |
|
"PL-MTEB--scifield" |
|
] |
|
], |
|
"embedding_head_config": { |
|
"dropout": 0.0, |
|
"n_dense": 1, |
|
"normalize_embeddings": false, |
|
"pool_type": "cls", |
|
"use_batch_norm": false, |
|
"use_layer_norm": false |
|
}, |
|
"hidden_size": 768, |
|
"mlm_head_config": { |
|
"dropout": 0.0, |
|
"n_dense": 1, |
|
"use_batch_norm": false, |
|
"use_layer_norm": true |
|
}, |
|
"model_type": "ILKT", |
|
"torch_dtype": "float32", |
|
"transformers_version": "4.41.2" |
|
} |
|
|