hieule commited on
Commit
30c8836
1 Parent(s): 158afb4

debeberta train withnew spm tokenizer

Browse files
Files changed (4) hide show
  1. config.json +31 -0
  2. pytorch_model.bin +3 -0
  3. tokenizer.json +0 -0
  4. tokenizer_config.json +2 -0
config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "checkpoints/deberta-vie-small_spm-3",
3
+ "architectures": [
4
+ "DebertaV2ForMaskedLM"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 2,
8
+ "eos_token_id": 3,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "layer_norm_eps": 1e-07,
15
+ "max_position_embeddings": 512,
16
+ "max_relative_positions": -1,
17
+ "model_type": "deberta-v2",
18
+ "num_attention_heads": 12,
19
+ "num_hidden_layers": 6,
20
+ "pad_token_id": 0,
21
+ "pooler_dropout": 0,
22
+ "pooler_hidden_act": "gelu",
23
+ "pooler_hidden_size": 768,
24
+ "pos_att_type": null,
25
+ "position_biased_input": true,
26
+ "relative_attention": false,
27
+ "torch_dtype": "float32",
28
+ "transformers_version": "4.22.1",
29
+ "type_vocab_size": 0,
30
+ "vocab_size": 45000
31
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37d732e29b13b620fbe04738f3949972681eb7eca7acdd2724e0c365813049fc
3
+ size 312514285
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json CHANGED
@@ -4,9 +4,11 @@
4
  "do_lower_case": false,
5
  "eos_token": "[EOS]",
6
  "mask_token": "[MASK]",
 
7
  "pad_token": "[PAD]",
8
  "sep_token": "[SEP]",
9
  "sp_model_kwargs": {},
 
10
  "split_by_punct": false,
11
  "tokenizer_class": "DebertaV2Tokenizer",
12
  "unk_token": "[UNK]"
 
4
  "do_lower_case": false,
5
  "eos_token": "[EOS]",
6
  "mask_token": "[MASK]",
7
+ "name_or_path": "checkpoints/deberta-vie-small_spm-3",
8
  "pad_token": "[PAD]",
9
  "sep_token": "[SEP]",
10
  "sp_model_kwargs": {},
11
+ "special_tokens_map_file": "/home/viet/.cache/huggingface/hub/models--hieule--spm-vie-deberta/snapshots/158afb4892bb8ed8ccdcfe9e60aa9a410d6a5325/special_tokens_map.json",
12
  "split_by_punct": false,
13
  "tokenizer_class": "DebertaV2Tokenizer",
14
  "unk_token": "[UNK]"