| { |
| "architectures": [ |
| "StrataBertForSequenceClassification" |
| ], |
| "attention_dropout": 0.0, |
| "auto_map": { |
| "AutoConfig": "configuration_stratabert.StrataBertConfig", |
| "AutoModel": "modeling_stratabert.StrataBertModel", |
| "AutoModelForMaskedLM": "modeling_stratabert.StrataBertForMaskedLM", |
| "AutoModelForSequenceClassification": "modeling_stratabert.StrataBertForSequenceClassification", |
| "AutoModelForTokenClassification": "modeling_stratabert.StrataBertForTokenClassification" |
| }, |
| "bidirectional_ssm": true, |
| "bos_token_id": null, |
| "classifier_dropout": 0.0, |
| "cls_token_id": 101, |
| "dtype": "float32", |
| "eos_token_id": null, |
| "global_attention_interval": 4, |
| "hidden_activation": "gelu", |
| "hidden_dropout": 0.0, |
| "hidden_size": 48, |
| "id2label": { |
| "0": "LABEL_0", |
| "1": "LABEL_1", |
| "2": "LABEL_2", |
| "3": "LABEL_3" |
| }, |
| "initializer_range": 0.02, |
| "intermediate_size": 96, |
| "label2id": { |
| "LABEL_0": 0, |
| "LABEL_1": 1, |
| "LABEL_2": 2, |
| "LABEL_3": 3 |
| }, |
| "layer_norm_eps": 1e-05, |
| "layer_types": [ |
| "global_attention", |
| "ssm", |
| "local_attention" |
| ], |
| "local_attention_window": 16, |
| "max_position_embeddings": 128, |
| "model_type": "stratabert", |
| "norm_eps": 1e-05, |
| "norm_type": "rmsnorm", |
| "num_attention_heads": 4, |
| "num_hidden_layers": 3, |
| "pad_token_id": 0, |
| "pooling_type": "hybrid_masked", |
| "sep_token_id": 102, |
| "ssm_conv_kernel": 4, |
| "ssm_expand": 2, |
| "ssm_fuse": "gated_sum", |
| "ssm_reset_on_pad": true, |
| "ssm_reset_on_segment": true, |
| "ssm_state_size": 24, |
| "ssm_type": "fallback_masked_scan", |
| "transformers_version": "5.11.0", |
| "type_vocab_size": 2, |
| "use_compile_safe_ssm": false, |
| "use_flash_attention": false, |
| "use_padding_free_attention": false, |
| "vocab_size": 50368 |
| } |
|
|
