Risheeb's picture
Training in progress, epoch 1
37f546a verified
{
"architectures": [
"ModernBertForTokenClassification"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 50281,
"classifier_activation": "gelu",
"classifier_bias": false,
"classifier_dropout": 0.0,
"classifier_pooling": "mean",
"cls_token_id": 50281,
"decoder_bias": true,
"deterministic_flash_attn": false,
"embedding_dropout": 0.0,
"eos_token_id": 50282,
"global_attn_every_n_layers": 3,
"global_rope_theta": 160000.0,
"gradient_checkpointing": false,
"hidden_activation": "gelu",
"hidden_size": 1024,
"id2label": {
"0": "O",
"1": "B-Concurrent",
"2": "B-Consecutive",
"3": "B-Dismissal",
"4": "B-Fine",
"5": "B-Imprisonment",
"6": "B-Other",
"7": "B-Probation",
"8": "B-Revocation",
"9": "B-Waived",
"10": "I-Concurrent",
"11": "I-Consecutive",
"12": "I-Dismissal",
"13": "I-Fine",
"14": "I-Imprisonment",
"15": "I-Other",
"16": "I-Probation",
"17": "I-Revocation",
"18": "I-Waived",
"19": "amounts",
"20": "assessment",
"21": "confinement",
"22": "consequences",
"23": "judgment",
"24": "registration",
"25": "requirement",
"26": "restrictions",
"27": "served",
"28": "service",
"29": "supervised",
"30": "time",
"31": "treatment",
"32": "unsupervised"
},
"initializer_cutoff_factor": 2.0,
"initializer_range": 0.02,
"intermediate_size": 2624,
"label2id": {
"B-Concurrent": 1,
"B-Consecutive": 2,
"B-Dismissal": 3,
"B-Fine": 4,
"B-Imprisonment": 5,
"B-Other": 6,
"B-Probation": 7,
"B-Revocation": 8,
"B-Waived": 9,
"I-Concurrent": 10,
"I-Consecutive": 11,
"I-Dismissal": 12,
"I-Fine": 13,
"I-Imprisonment": 14,
"I-Other": 15,
"I-Probation": 16,
"I-Revocation": 17,
"I-Waived": 18,
"O": 0,
"amounts": 19,
"assessment": 20,
"confinement": 21,
"consequences": 22,
"judgment": 23,
"registration": 24,
"requirement": 25,
"restrictions": 26,
"served": 27,
"service": 28,
"supervised": 29,
"time": 30,
"treatment": 31,
"unsupervised": 32
},
"layer_norm_eps": 1e-05,
"local_attention": 128,
"local_rope_theta": 10000.0,
"max_position_embeddings": 8192,
"mlp_bias": false,
"mlp_dropout": 0.0,
"model_type": "modernbert",
"norm_bias": false,
"norm_eps": 1e-05,
"num_attention_heads": 16,
"num_hidden_layers": 28,
"pad_token_id": 50283,
"position_embedding_type": "absolute",
"repad_logits_with_grad": false,
"sep_token_id": 50282,
"sparse_pred_ignore_index": -100,
"sparse_prediction": false,
"torch_dtype": "float32",
"transformers_version": "4.51.1",
"vocab_size": 50368
}