Shubham Toshniwal commited on
Commit
27d97a0
1 Parent(s): 1e27217

Upload config.json

Browse files
Files changed (1) hide show
  1. config.json +57 -0
config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/share/data/speech/shtoshni/research/fast-coref/models/coref_joint_11c4b92806f1db7437083841caffe8da/doc_encoder",
3
+ "architectures": [
4
+ "LongformerModel"
5
+ ],
6
+ "attention_mode": "longformer",
7
+ "attention_probs_dropout_prob": 0.1,
8
+ "attention_window": [
9
+ 512,
10
+ 512,
11
+ 512,
12
+ 512,
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512,
20
+ 512,
21
+ 512,
22
+ 512,
23
+ 512,
24
+ 512,
25
+ 512,
26
+ 512,
27
+ 512,
28
+ 512,
29
+ 512,
30
+ 512,
31
+ 512,
32
+ 512
33
+ ],
34
+ "bos_token_id": 0,
35
+ "classifier_dropout": null,
36
+ "eos_token_id": 2,
37
+ "gradient_checkpointing": false,
38
+ "hidden_act": "gelu",
39
+ "hidden_dropout_prob": 0.1,
40
+ "hidden_size": 1024,
41
+ "ignore_attention_mask": false,
42
+ "initializer_range": 0.02,
43
+ "intermediate_size": 4096,
44
+ "layer_norm_eps": 1e-05,
45
+ "max_position_embeddings": 4098,
46
+ "model_type": "longformer",
47
+ "num_attention_heads": 16,
48
+ "num_hidden_layers": 24,
49
+ "pad_token_id": 1,
50
+ "position_embedding_type": "absolute",
51
+ "sep_token_id": 2,
52
+ "torch_dtype": "float32",
53
+ "transformers_version": "4.11.3",
54
+ "type_vocab_size": 1,
55
+ "use_cache": true,
56
+ "vocab_size": 50267
57
+ }