SophieTr commited on
Commit
5f0471d
1 Parent(s): b75eebe
Files changed (2) hide show
  1. config.json +29 -15
  2. pytorch_model.bin +2 -2
config.json CHANGED
@@ -1,44 +1,58 @@
1
  {
2
- "_name_or_path": "google/bigbird-pegasus-large-arxiv",
3
- "activation_dropout": 0.0,
4
- "activation_function": "gelu_new",
 
 
5
  "architectures": [
6
  "PegasusModel"
7
  ],
8
- "attention_dropout": 0.0,
9
- "attention_type": "block_sparse",
10
- "block_size": 64,
11
- "bos_token_id": 2,
12
  "classifier_dropout": 0.0,
13
  "d_model": 1024,
14
  "decoder_attention_heads": 16,
15
  "decoder_ffn_dim": 4096,
16
  "decoder_layerdrop": 0.0,
17
  "decoder_layers": 16,
18
- "decoder_start_token_id": 2,
 
19
  "dropout": 0.1,
20
  "encoder_attention_heads": 16,
21
  "encoder_ffn_dim": 4096,
22
  "encoder_layerdrop": 0.0,
23
  "encoder_layers": 16,
24
  "eos_token_id": 1,
 
 
25
  "forced_eos_token_id": 1,
26
  "gradient_checkpointing": false,
 
 
 
 
 
27
  "init_std": 0.02,
28
  "is_encoder_decoder": true,
29
- "length_penalty": 0.8,
30
- "max_length": 256,
31
- "max_position_embeddings": 4096,
 
 
 
 
 
32
  "model_type": "pegasus",
33
- "num_beams": 5,
 
 
34
  "num_hidden_layers": 16,
35
- "num_random_blocks": 3,
36
  "pad_token_id": 0,
37
  "scale_embedding": true,
38
- "tokenizer_class": "PegasusTokenizer",
39
  "torch_dtype": "float32",
40
  "transformers_version": "4.15.0",
41
- "use_bias": false,
42
  "use_cache": true,
43
  "vocab_size": 96103
44
  }
 
1
  {
2
+ "_name_or_path": "google/pegasus-xsum",
3
+ "activation_dropout": 0.1,
4
+ "activation_function": "relu",
5
+ "add_bias_logits": false,
6
+ "add_final_layer_norm": true,
7
  "architectures": [
8
  "PegasusModel"
9
  ],
10
+ "attention_dropout": 0.1,
11
+ "bos_token_id": 0,
12
+ "classif_dropout": 0.0,
 
13
  "classifier_dropout": 0.0,
14
  "d_model": 1024,
15
  "decoder_attention_heads": 16,
16
  "decoder_ffn_dim": 4096,
17
  "decoder_layerdrop": 0.0,
18
  "decoder_layers": 16,
19
+ "decoder_start_token_id": 0,
20
+ "do_blenderbot_90_layernorm": false,
21
  "dropout": 0.1,
22
  "encoder_attention_heads": 16,
23
  "encoder_ffn_dim": 4096,
24
  "encoder_layerdrop": 0.0,
25
  "encoder_layers": 16,
26
  "eos_token_id": 1,
27
+ "extra_pos_embeddings": 0,
28
+ "force_bos_token_to_be_generated": false,
29
  "forced_eos_token_id": 1,
30
  "gradient_checkpointing": false,
31
+ "id2label": {
32
+ "0": "LABEL_0",
33
+ "1": "LABEL_1",
34
+ "2": "LABEL_2"
35
+ },
36
  "init_std": 0.02,
37
  "is_encoder_decoder": true,
38
+ "label2id": {
39
+ "LABEL_0": 0,
40
+ "LABEL_1": 1,
41
+ "LABEL_2": 2
42
+ },
43
+ "length_penalty": 0.6,
44
+ "max_length": 64,
45
+ "max_position_embeddings": 512,
46
  "model_type": "pegasus",
47
+ "normalize_before": true,
48
+ "normalize_embedding": false,
49
+ "num_beams": 8,
50
  "num_hidden_layers": 16,
 
51
  "pad_token_id": 0,
52
  "scale_embedding": true,
53
+ "static_position_embeddings": true,
54
  "torch_dtype": "float32",
55
  "transformers_version": "4.15.0",
 
56
  "use_cache": true,
57
  "vocab_size": 96103
58
  }
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ac5570d940a0dba5ed3b5e146f08610a820e2bc0e7cdbcd1d600a7e2c4460974
3
- size 2308600115
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eea24ef4477b4819aa83f02f402ebeb7e8eb0e3000e11d54aae748e5f27ac657
3
+ size 2279230451