Upload tokenizer_config.json with huggingface_hub
Browse files- tokenizer_config.json +1 -44
tokenizer_config.json
CHANGED
@@ -25,59 +25,16 @@
|
|
25 |
"rstrip": false,
|
26 |
"single_word": false,
|
27 |
"special": true
|
28 |
-
},
|
29 |
-
"32007": {
|
30 |
-
"content": "β<PRE>",
|
31 |
-
"lstrip": false,
|
32 |
-
"normalized": false,
|
33 |
-
"rstrip": false,
|
34 |
-
"single_word": false,
|
35 |
-
"special": true
|
36 |
-
},
|
37 |
-
"32008": {
|
38 |
-
"content": "β<SUF>",
|
39 |
-
"lstrip": false,
|
40 |
-
"normalized": false,
|
41 |
-
"rstrip": false,
|
42 |
-
"single_word": false,
|
43 |
-
"special": true
|
44 |
-
},
|
45 |
-
"32009": {
|
46 |
-
"content": "β<MID>",
|
47 |
-
"lstrip": false,
|
48 |
-
"normalized": false,
|
49 |
-
"rstrip": false,
|
50 |
-
"single_word": false,
|
51 |
-
"special": true
|
52 |
-
},
|
53 |
-
"32010": {
|
54 |
-
"content": "β<EOT>",
|
55 |
-
"lstrip": false,
|
56 |
-
"normalized": false,
|
57 |
-
"rstrip": false,
|
58 |
-
"single_word": false,
|
59 |
-
"special": true
|
60 |
}
|
61 |
},
|
62 |
-
"additional_special_tokens": [
|
63 |
-
"β<PRE>",
|
64 |
-
"β<MID>",
|
65 |
-
"β<SUF>",
|
66 |
-
"β<EOT>"
|
67 |
-
],
|
68 |
"bos_token": "<s>",
|
69 |
"clean_up_tokenization_spaces": false,
|
70 |
"eos_token": "</s>",
|
71 |
-
"eot_token": "β<EOT>",
|
72 |
-
"fill_token": "<FILL_ME>",
|
73 |
"legacy": null,
|
74 |
-
"middle_token": "β<MID>",
|
75 |
"model_max_length": 1000000000000000019884624838656,
|
76 |
"pad_token": "[PAD]",
|
77 |
-
"prefix_token": "β<PRE>",
|
78 |
"sp_model_kwargs": {},
|
79 |
-
"
|
80 |
-
"tokenizer_class": "CodeLlamaTokenizer",
|
81 |
"unk_token": "<unk>",
|
82 |
"use_default_system_prompt": false
|
83 |
}
|
|
|
25 |
"rstrip": false,
|
26 |
"single_word": false,
|
27 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
28 |
}
|
29 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
30 |
"bos_token": "<s>",
|
31 |
"clean_up_tokenization_spaces": false,
|
32 |
"eos_token": "</s>",
|
|
|
|
|
33 |
"legacy": null,
|
|
|
34 |
"model_max_length": 1000000000000000019884624838656,
|
35 |
"pad_token": "[PAD]",
|
|
|
36 |
"sp_model_kwargs": {},
|
37 |
+
"tokenizer_class": "LlamaTokenizer",
|
|
|
38 |
"unk_token": "<unk>",
|
39 |
"use_default_system_prompt": false
|
40 |
}
|