Commit
•
ceeb875
1
Parent(s):
34722f0
Upload tokenizer
Browse files- added_tokens.json +0 -2
- tokenizer.json +0 -18
- tokenizer_config.json +1 -19
added_tokens.json
CHANGED
@@ -1,6 +1,4 @@
|
|
1 |
{
|
2 |
"<|end_of_turn|>": 32000,
|
3 |
-
"<|im_end|>": 32002,
|
4 |
-
"<|im_start|>": 32003,
|
5 |
"<|pad_0|>": 32001
|
6 |
}
|
|
|
1 |
{
|
2 |
"<|end_of_turn|>": 32000,
|
|
|
|
|
3 |
"<|pad_0|>": 32001
|
4 |
}
|
tokenizer.json
CHANGED
@@ -47,24 +47,6 @@
|
|
47 |
"rstrip": false,
|
48 |
"normalized": false,
|
49 |
"special": true
|
50 |
-
},
|
51 |
-
{
|
52 |
-
"id": 32002,
|
53 |
-
"content": "<|im_end|>",
|
54 |
-
"single_word": false,
|
55 |
-
"lstrip": false,
|
56 |
-
"rstrip": false,
|
57 |
-
"normalized": false,
|
58 |
-
"special": true
|
59 |
-
},
|
60 |
-
{
|
61 |
-
"id": 32003,
|
62 |
-
"content": "<|im_start|>",
|
63 |
-
"single_word": false,
|
64 |
-
"lstrip": false,
|
65 |
-
"rstrip": false,
|
66 |
-
"normalized": false,
|
67 |
-
"special": true
|
68 |
}
|
69 |
],
|
70 |
"normalizer": {
|
|
|
47 |
"rstrip": false,
|
48 |
"normalized": false,
|
49 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
50 |
}
|
51 |
],
|
52 |
"normalizer": {
|
tokenizer_config.json
CHANGED
@@ -39,22 +39,6 @@
|
|
39 |
"rstrip": false,
|
40 |
"single_word": false,
|
41 |
"special": true
|
42 |
-
},
|
43 |
-
"32002": {
|
44 |
-
"content": "<|im_end|>",
|
45 |
-
"lstrip": false,
|
46 |
-
"normalized": false,
|
47 |
-
"rstrip": false,
|
48 |
-
"single_word": false,
|
49 |
-
"special": true
|
50 |
-
},
|
51 |
-
"32003": {
|
52 |
-
"content": "<|im_start|>",
|
53 |
-
"lstrip": false,
|
54 |
-
"normalized": false,
|
55 |
-
"rstrip": false,
|
56 |
-
"single_word": false,
|
57 |
-
"special": true
|
58 |
}
|
59 |
},
|
60 |
"additional_special_tokens": [
|
@@ -66,9 +50,7 @@
|
|
66 |
],
|
67 |
"bos_token": "<s>",
|
68 |
"clean_up_tokenization_spaces": false,
|
69 |
-
"device_map":
|
70 |
-
"": "cuda"
|
71 |
-
},
|
72 |
"eos_token": "<|end_of_turn|>",
|
73 |
"legacy": true,
|
74 |
"model_max_length": 16384,
|
|
|
39 |
"rstrip": false,
|
40 |
"single_word": false,
|
41 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
42 |
}
|
43 |
},
|
44 |
"additional_special_tokens": [
|
|
|
50 |
],
|
51 |
"bos_token": "<s>",
|
52 |
"clean_up_tokenization_spaces": false,
|
53 |
+
"device_map": "auto",
|
|
|
|
|
54 |
"eos_token": "<|end_of_turn|>",
|
55 |
"legacy": true,
|
56 |
"model_max_length": 16384,
|