elucidator8918 commited on
Commit
d7dc351
·
verified ·
1 Parent(s): 9a3424c

Upload config

Browse files
Files changed (1) hide show
  1. config.json +66 -3
config.json CHANGED
@@ -1,4 +1,67 @@
1
  {
2
- "model_path": "Alibaba-NLP/gte-modernbert-base",
3
- "num_labels": 8
4
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  {
2
+ "_name_or_path": "Alibaba-NLP/gte-modernbert-base",
3
+ "architectures": [
4
+ "ModernBertModel"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 50281,
9
+ "classifier_activation": "gelu",
10
+ "classifier_bias": false,
11
+ "classifier_dropout": 0.0,
12
+ "classifier_pooling": "mean",
13
+ "cls_token_id": 50281,
14
+ "decoder_bias": true,
15
+ "deterministic_flash_attn": false,
16
+ "embedding_dropout": 0.0,
17
+ "eos_token_id": 50282,
18
+ "global_attn_every_n_layers": 3,
19
+ "global_rope_theta": 160000.0,
20
+ "gradient_checkpointing": false,
21
+ "hidden_activation": "gelu",
22
+ "hidden_size": 768,
23
+ "id2label": {
24
+ "0": "LABEL_0",
25
+ "1": "LABEL_1",
26
+ "2": "LABEL_2",
27
+ "3": "LABEL_3",
28
+ "4": "LABEL_4",
29
+ "5": "LABEL_5",
30
+ "6": "LABEL_6",
31
+ "7": "LABEL_7"
32
+ },
33
+ "initializer_cutoff_factor": 2.0,
34
+ "initializer_range": 0.02,
35
+ "intermediate_size": 1152,
36
+ "label2id": {
37
+ "LABEL_0": 0,
38
+ "LABEL_1": 1,
39
+ "LABEL_2": 2,
40
+ "LABEL_3": 3,
41
+ "LABEL_4": 4,
42
+ "LABEL_5": 5,
43
+ "LABEL_6": 6,
44
+ "LABEL_7": 7
45
+ },
46
+ "layer_norm_eps": 1e-05,
47
+ "local_attention": 128,
48
+ "local_rope_theta": 10000.0,
49
+ "max_position_embeddings": 8192,
50
+ "mlp_bias": false,
51
+ "mlp_dropout": 0.0,
52
+ "model_type": "modernbert",
53
+ "norm_bias": false,
54
+ "norm_eps": 1e-05,
55
+ "num_attention_heads": 12,
56
+ "num_hidden_layers": 22,
57
+ "pad_token_id": 50283,
58
+ "position_embedding_type": "absolute",
59
+ "reference_compile": null,
60
+ "repad_logits_with_grad": false,
61
+ "sep_token_id": 50282,
62
+ "sparse_pred_ignore_index": -100,
63
+ "sparse_prediction": false,
64
+ "torch_dtype": "float16",
65
+ "transformers_version": "4.48.1",
66
+ "vocab_size": 50368
67
+ }