danielsaggau/scotus_max_linear_frozen

1{2  "_name_or_path": "danielsaggau/scotus_max_pool",3  "architectures": [4    "LongformerForSequenceClassification"5  ],6  "attention_mode": "longformer",7  "attention_probs_dropout_prob": 0.1,8  "attention_window": [9    128,10    128,11    128,12    128,13    128,14    12815  ],16  "bos_token_id": 0,17  "classifier_dropout": null,18  "cls_token_id": 101,19  "eos_token_id": null,20  "gradient_checkpointing": false,21  "hidden_act": "gelu",22  "hidden_dropout_prob": 0.1,23  "hidden_size": 512,24  "id2label": {25    "0": "LABEL_0",26    "1": "LABEL_1",27    "2": "LABEL_2",28    "3": "LABEL_3",29    "4": "LABEL_4",30    "5": "LABEL_5",31    "6": "LABEL_6",32    "7": "LABEL_7",33    "8": "LABEL_8",34    "9": "LABEL_9",35    "10": "LABEL_10",36    "11": "LABEL_11",37    "12": "LABEL_12",38    "13": "LABEL_13"39  },40  "ignore_attention_mask": false,41  "initializer_range": 0.02,42  "intermediate_size": 2048,43  "label2id": {44    "LABEL_0": 0,45    "LABEL_1": 1,46    "LABEL_10": 10,47    "LABEL_11": 11,48    "LABEL_12": 12,49    "LABEL_13": 13,50    "LABEL_2": 2,51    "LABEL_3": 3,52    "LABEL_4": 4,53    "LABEL_5": 5,54    "LABEL_6": 6,55    "LABEL_7": 7,56    "LABEL_8": 8,57    "LABEL_9": 958  },59  "layer_norm_eps": 1e-05,60  "max_position_embeddings": 4098,61  "model_max_length": 4096,62  "model_type": "longformer",63  "num_attention_heads": 8,64  "num_hidden_layers": 6,65  "onnx_export": false,66  "pad_token_id": 0,67  "position_embedding_type": "absolute",68  "problem_type": "single_label_classification",69  "sep_token_id": 102,70  "torch_dtype": "float32",71  "transformers_version": "4.23.1",72  "type_vocab_size": 2,73  "use_cache": true,74  "vocab_size": 3052275}76