config: upload initial version
Browse files- config.json +79 -0
config.json
ADDED
@@ -0,0 +1,79 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "/home/stefan/Repositories/NeoBERT",
|
3 |
+
"architectures": [
|
4 |
+
"NeoBERTForTokenClassification"
|
5 |
+
],
|
6 |
+
"auto_map": {
|
7 |
+
"AutoConfig": "model.NeoBERTConfig",
|
8 |
+
"AutoModel": "model.NeoBERT",
|
9 |
+
"AutoModelForMaskedLM": "model.NeoBERTLMHead",
|
10 |
+
"AutoModelForSequenceClassification": "model.NeoBERTForSequenceClassification",
|
11 |
+
"AutoModelForTokenClassification": "model.NeoBERTForTokenClassification"
|
12 |
+
},
|
13 |
+
"classifier_init_range": 0.02,
|
14 |
+
"decoder_init_range": 0.02,
|
15 |
+
"dim_head": 64,
|
16 |
+
"embedding_init_range": 0.02,
|
17 |
+
"finetuning_task": "ner",
|
18 |
+
"hidden_size": 768,
|
19 |
+
"id2label": {
|
20 |
+
"0": "O",
|
21 |
+
"1": "B-PER",
|
22 |
+
"2": "I-PER",
|
23 |
+
"3": "B-ORG",
|
24 |
+
"4": "I-ORG",
|
25 |
+
"5": "B-LOC",
|
26 |
+
"6": "I-LOC",
|
27 |
+
"7": "B-MISC",
|
28 |
+
"8": "I-MISC"
|
29 |
+
},
|
30 |
+
"intermediate_size": 3072,
|
31 |
+
"kwargs": {
|
32 |
+
"_commit_hash": null,
|
33 |
+
"architectures": [
|
34 |
+
"NeoBERTLMHead"
|
35 |
+
],
|
36 |
+
"attn_implementation": null,
|
37 |
+
"auto_map": {
|
38 |
+
"AutoConfig": "model.NeoBERTConfig",
|
39 |
+
"AutoModel": "model.NeoBERT",
|
40 |
+
"AutoModelForMaskedLM": "model.NeoBERTLMHead",
|
41 |
+
"AutoModelForSequenceClassification": "model.NeoBERTForSequenceClassification",
|
42 |
+
"AutoModelForTokenClassification": "model.NeoBERTForTokenClassification"
|
43 |
+
},
|
44 |
+
"classifier_init_range": 0.02,
|
45 |
+
"dim_head": 64,
|
46 |
+
"kwargs": {
|
47 |
+
"classifier_init_range": 0.02,
|
48 |
+
"pretrained_model_name_or_path": "google-bert/bert-base-uncased",
|
49 |
+
"trust_remote_code": true
|
50 |
+
},
|
51 |
+
"model_type": "neobert",
|
52 |
+
"pretrained_model_name_or_path": "google-bert/bert-base-uncased",
|
53 |
+
"torch_dtype": "float32",
|
54 |
+
"transformers_version": "4.48.2",
|
55 |
+
"trust_remote_code": true
|
56 |
+
},
|
57 |
+
"label2id": {
|
58 |
+
"B-LOC": 5,
|
59 |
+
"B-MISC": 7,
|
60 |
+
"B-ORG": 3,
|
61 |
+
"B-PER": 1,
|
62 |
+
"I-LOC": 6,
|
63 |
+
"I-MISC": 8,
|
64 |
+
"I-ORG": 4,
|
65 |
+
"I-PER": 2,
|
66 |
+
"O": 0
|
67 |
+
},
|
68 |
+
"max_length": 4096,
|
69 |
+
"model_type": "neobert",
|
70 |
+
"norm_eps": 1e-05,
|
71 |
+
"num_attention_heads": 12,
|
72 |
+
"num_hidden_layers": 28,
|
73 |
+
"pad_token_id": 0,
|
74 |
+
"pretrained_model_name_or_path": "google-bert/bert-base-uncased",
|
75 |
+
"torch_dtype": "float32",
|
76 |
+
"transformers_version": "4.49.0",
|
77 |
+
"trust_remote_code": true,
|
78 |
+
"vocab_size": 30522
|
79 |
+
}
|