| { |
| "_name_or_path": "ibm/MoLFormer-XL-both-10pct", |
| "architectures": [ |
| "MolformerModel" |
| ], |
| "auto_map": { |
| "AutoConfig": "ibm/MoLFormer-XL-both-10pct--configuration_molformer.MolformerConfig", |
| "AutoModel": "ibm/MoLFormer-XL-both-10pct--modeling_molformer.MolformerModel", |
| "AutoModelForMaskedLM": "ibm/MoLFormer-XL-both-10pct--modeling_molformer.MolformerForMaskedLM", |
| "AutoModelForSequenceClassification": "ibm/MoLFormer-XL-both-10pct--modeling_molformer.MolformerForSequenceClassification" |
| }, |
| "classifier_dropout_prob": null, |
| "classifier_skip_connection": true, |
| "deterministic_eval": true, |
| "embedding_dropout_prob": 0.2, |
| "feature_map_kernel": "relu", |
| "hidden_act": "gelu", |
| "hidden_dropout_prob": 0.1, |
| "hidden_size": 768, |
| "initializer_range": 0.02, |
| "intermediate_size": 768, |
| "layer_norm_eps": 1e-12, |
| "linear_attention_eps": 1e-06, |
| "max_position_embeddings": 202, |
| "model_type": "molformer", |
| "num_attention_heads": 12, |
| "num_hidden_layers": 12, |
| "num_random_features": 32, |
| "pad_token_id": 2, |
| "tie_word_embeddings": false, |
| "torch_dtype": "float32", |
| "transformers_version": "4.46.3", |
| "vocab_size": 2362 |
| } |
|
|
