Hugging Face's logo

JackFram
/
llama-160m

Text Generation
Transformers
PyTorch
Safetensors
English
llama
text-generation-inference
Inference Endpoints
Model card Files Community
15
goliaro commited on
Commit
53e9c9a
•
1 Parent(s): a57a4d3

align tokenizer with llama 2

Browse files
Files changed (5) hide show
  1. config.json +2 -2
  2. generation_config.json +3 -4
  3. special_tokens_map.json +3 -3
  4. tokenizer.json +10 -10
  5. tokenizer_config.json +6 -5
config.json CHANGED
@@ -13,11 +13,11 @@
13
  "model_type": "llama",
14
  "num_attention_heads": 12,
15
  "num_hidden_layers": 12,
16
- "pad_token_id": 1,
17
  "rms_norm_eps": 1e-06,
18
  "tie_word_embeddings": false,
19
  "torch_dtype": "float32",
20
- "transformers_version": "4.30.0.dev0",
21
  "use_cache": true,
22
  "vocab_size": 32000
23
  }
 
13
  "model_type": "llama",
14
  "num_attention_heads": 12,
15
  "num_hidden_layers": 12,
16
+ "num_key_value_heads": 12,
17
  "rms_norm_eps": 1e-06,
18
  "tie_word_embeddings": false,
19
  "torch_dtype": "float32",
20
+ "transformers_version": "4.31.0.dev0",
21
  "use_cache": true,
22
  "vocab_size": 32000
23
  }
generation_config.json CHANGED
@@ -1,7 +1,6 @@
1
  {
2
- "_from_model_config": true,
3
- "bos_token_id": 0,
4
  "eos_token_id": 2,
5
- "pad_token_id": 1,
6
- "transformers_version": "4.30.0.dev0"
7
  }
 
1
  {
2
+ "bos_token_id": 1,
 
3
  "eos_token_id": 2,
4
+ "pad_token_id": 0,
5
+ "transformers_version": "4.31.0.dev0"
6
  }
special_tokens_map.json CHANGED
@@ -2,21 +2,21 @@
2
  "bos_token": {
3
  "content": "<s>",
4
  "lstrip": false,
5
- "normalized": true,
6
  "rstrip": false,
7
  "single_word": false
8
  },
9
  "eos_token": {
10
  "content": "</s>",
11
  "lstrip": false,
12
- "normalized": true,
13
  "rstrip": false,
14
  "single_word": false
15
  },
16
  "unk_token": {
17
  "content": "<unk>",
18
  "lstrip": false,
19
- "normalized": true,
20
  "rstrip": false,
21
  "single_word": false
22
  }
 
2
  "bos_token": {
3
  "content": "<s>",
4
  "lstrip": false,
5
+ "normalized": false,
6
  "rstrip": false,
7
  "single_word": false
8
  },
9
  "eos_token": {
10
  "content": "</s>",
11
  "lstrip": false,
12
+ "normalized": false,
13
  "rstrip": false,
14
  "single_word": false
15
  },
16
  "unk_token": {
17
  "content": "<unk>",
18
  "lstrip": false,
19
+ "normalized": false,
20
  "rstrip": false,
21
  "single_word": false
22
  }
tokenizer.json CHANGED
@@ -9,7 +9,7 @@
9
  "single_word": false,
10
  "lstrip": false,
11
  "rstrip": false,
12
- "normalized": true,
13
  "special": true
14
  },
15
  {
@@ -18,7 +18,7 @@
18
  "single_word": false,
19
  "lstrip": false,
20
  "rstrip": false,
21
- "normalized": true,
22
  "special": true
23
  },
24
  {
@@ -27,7 +27,7 @@
27
  "single_word": false,
28
  "lstrip": false,
29
  "rstrip": false,
30
- "normalized": true,
31
  "special": true
32
  }
33
  ],
@@ -53,7 +53,7 @@
53
  "single": [
54
  {
55
  "SpecialToken": {
56
- "id": "",
57
  "type_id": 0
58
  }
59
  },
@@ -67,7 +67,7 @@
67
  "pair": [
68
  {
69
  "SpecialToken": {
70
- "id": "",
71
  "type_id": 0
72
  }
73
  },
@@ -79,7 +79,7 @@
79
  },
80
  {
81
  "SpecialToken": {
82
- "id": "",
83
  "type_id": 1
84
  }
85
  },
@@ -91,13 +91,13 @@
91
  }
92
  ],
93
  "special_tokens": {
94
- "": {
95
- "id": "",
96
  "ids": [
97
- 0
98
  ],
99
  "tokens": [
100
- ""
101
  ]
102
  }
103
  }
 
9
  "single_word": false,
10
  "lstrip": false,
11
  "rstrip": false,
12
+ "normalized": false,
13
  "special": true
14
  },
15
  {
 
18
  "single_word": false,
19
  "lstrip": false,
20
  "rstrip": false,
21
+ "normalized": false,
22
  "special": true
23
  },
24
  {
 
27
  "single_word": false,
28
  "lstrip": false,
29
  "rstrip": false,
30
+ "normalized": false,
31
  "special": true
32
  }
33
  ],
 
53
  "single": [
54
  {
55
  "SpecialToken": {
56
+ "id": "<s>",
57
  "type_id": 0
58
  }
59
  },
 
67
  "pair": [
68
  {
69
  "SpecialToken": {
70
+ "id": "<s>",
71
  "type_id": 0
72
  }
73
  },
 
79
  },
80
  {
81
  "SpecialToken": {
82
+ "id": "<s>",
83
  "type_id": 1
84
  }
85
  },
 
91
  }
92
  ],
93
  "special_tokens": {
94
+ "<s>": {
95
+ "id": "<s>",
96
  "ids": [
97
+ 1
98
  ],
99
  "tokens": [
100
+ "<s>"
101
  ]
102
  }
103
  }
tokenizer_config.json CHANGED
@@ -5,7 +5,7 @@
5
  "__type": "AddedToken",
6
  "content": "<s>",
7
  "lstrip": false,
8
- "normalized": true,
9
  "rstrip": false,
10
  "single_word": false
11
  },
@@ -14,21 +14,22 @@
14
  "__type": "AddedToken",
15
  "content": "</s>",
16
  "lstrip": false,
17
- "normalized": true,
18
  "rstrip": false,
19
  "single_word": false
20
  },
 
21
  "model_max_length": 1000000000000000019884624838656,
22
  "pad_token": null,
 
23
  "sp_model_kwargs": {},
24
  "tokenizer_class": "LlamaTokenizer",
25
  "unk_token": {
26
  "__type": "AddedToken",
27
  "content": "<unk>",
28
  "lstrip": false,
29
- "normalized": true,
30
  "rstrip": false,
31
  "single_word": false
32
- },
33
- "use_fast": true
34
  }
 
5
  "__type": "AddedToken",
6
  "content": "<s>",
7
  "lstrip": false,
8
+ "normalized": false,
9
  "rstrip": false,
10
  "single_word": false
11
  },
 
14
  "__type": "AddedToken",
15
  "content": "</s>",
16
  "lstrip": false,
17
+ "normalized": false,
18
  "rstrip": false,
19
  "single_word": false
20
  },
21
+ "legacy": false,
22
  "model_max_length": 1000000000000000019884624838656,
23
  "pad_token": null,
24
+ "padding_side": "right",
25
  "sp_model_kwargs": {},
26
  "tokenizer_class": "LlamaTokenizer",
27
  "unk_token": {
28
  "__type": "AddedToken",
29
  "content": "<unk>",
30
  "lstrip": false,
31
+ "normalized": false,
32
  "rstrip": false,
33
  "single_word": false
34
+ }
 
35
  }