graf / encoder /config.json
cjdd3b's picture
Upload folder using huggingface_hub
1c0079f verified
Raw History Blame Contribute Delete
1.71 kB
{
"architectures": [
"ModernBertForTokenClassification"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 42889,
"causal_mask": false,
"classifier_activation": "gelu",
"classifier_bias": false,
"classifier_dropout": 0.0,
"classifier_pooling": "mean",
"cls_token_id": 42889,
"decoder_bias": true,
"deterministic_flash_attn": false,
"dtype": "float32",
"embedding_dropout": 0.0,
"eos_token_id": 42890,
"global_attn_every_n_layers": 3,
"gradient_checkpointing": false,
"hidden_activation": "gelu",
"hidden_size": 256,
"id2label": {
"0": "NO_BREAK",
"1": "BREAK"
},
"initializer_cutoff_factor": 2.0,
"initializer_range": 0.02,
"intermediate_size": 384,
"is_causal": false,
"label2id": {
"BREAK": 1,
"NO_BREAK": 0
},
"layer_norm_eps": 1e-05,
"layer_types": [
"full_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"full_attention"
],
"local_attention": 128,
"max_position_embeddings": 7999,
"mlp_bias": false,
"mlp_dropout": 0.0,
"model_type": "modernbert",
"norm_bias": false,
"norm_eps": 1e-05,
"num_attention_heads": 4,
"num_hidden_layers": 7,
"pad_token_id": 42891,
"position_embedding_type": "sans_pos",
"rope_parameters": {
"full_attention": {
"rope_theta": 160000.0,
"rope_type": "default"
},
"sliding_attention": {
"rope_theta": 160000.0,
"rope_type": "default"
}
},
"sep_token_id": 42890,
"sparse_pred_ignore_index": -100,
"sparse_prediction": false,
"tie_word_embeddings": true,
"transformers_version": "5.17.0",
"vocab_size": 42893
}