Commit
•
ac7e18c
1
Parent(s):
fbc057c
nvl27g-sulion-shards
Browse files- config.json +4 -4
- generation_config.json +1 -1
- pytorch_model-00001-of-00014.bin +1 -1
- pytorch_model-00002-of-00014.bin +1 -1
- pytorch_model-00003-of-00014.bin +1 -1
- pytorch_model-00004-of-00014.bin +1 -1
- pytorch_model-00005-of-00014.bin +1 -1
- pytorch_model-00006-of-00014.bin +1 -1
- pytorch_model-00007-of-00014.bin +1 -1
- pytorch_model-00008-of-00014.bin +1 -1
- pytorch_model-00009-of-00014.bin +1 -1
- pytorch_model-00010-of-00014.bin +1 -1
- pytorch_model-00011-of-00014.bin +1 -1
- pytorch_model-00012-of-00014.bin +1 -1
- pytorch_model-00013-of-00014.bin +1 -1
- pytorch_model-00014-of-00014.bin +1 -1
- special_tokens_map.json +9 -28
- tokenizer.json +32 -4
- tokenizer_config.json +35 -33
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"apply_residual_connection_post_layernorm": false,
|
4 |
"architectures": [
|
5 |
"CCBloomForCausalLM"
|
@@ -31,10 +31,10 @@
|
|
31 |
"skip_bias_add_qkv": false,
|
32 |
"slow_but_exact": false,
|
33 |
"tie_word_embeddings": false,
|
34 |
-
"torch_dtype": "
|
35 |
"transformers_version": "4.31.0",
|
36 |
"unk_token_id": 0,
|
37 |
"use_cache": true,
|
38 |
-
"use_flash_attn":
|
39 |
"vocab_size": 64000
|
40 |
-
}
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "checkpoint_final/",
|
3 |
"apply_residual_connection_post_layernorm": false,
|
4 |
"architectures": [
|
5 |
"CCBloomForCausalLM"
|
|
|
31 |
"skip_bias_add_qkv": false,
|
32 |
"slow_but_exact": false,
|
33 |
"tie_word_embeddings": false,
|
34 |
+
"torch_dtype": "float16",
|
35 |
"transformers_version": "4.31.0",
|
36 |
"unk_token_id": 0,
|
37 |
"use_cache": true,
|
38 |
+
"use_flash_attn": true,
|
39 |
"vocab_size": 64000
|
40 |
+
}
|
generation_config.json
CHANGED
@@ -2,6 +2,6 @@
|
|
2 |
"_from_model_config": true,
|
3 |
"bos_token_id": 1,
|
4 |
"eos_token_id": 2,
|
5 |
-
"pad_token_id":
|
6 |
"transformers_version": "4.31.0"
|
7 |
}
|
|
|
2 |
"_from_model_config": true,
|
3 |
"bos_token_id": 1,
|
4 |
"eos_token_id": 2,
|
5 |
+
"pad_token_id": 0,
|
6 |
"transformers_version": "4.31.0"
|
7 |
}
|
pytorch_model-00001-of-00014.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 927086401
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:97d4c7a9edcca08efe89e9dfdc606b0dfc4e366d3a06f08583022ce1d11f493d
|
3 |
size 927086401
|
pytorch_model-00002-of-00014.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 939796331
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b46f911795cbc625b2dec55ddbb6caffffa3756fb60854e447f7826bbacb8390
|
3 |
size 939796331
|
pytorch_model-00003-of-00014.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 939778659
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:aec035b6936f12656a523132caf4e41f5c3b5352bcd0c723207bae0418aca330
|
3 |
size 939778659
|
pytorch_model-00004-of-00014.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 939771079
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5d679e368e71d4a5d4624b14603cd3c6770b629ea8e0f985661fa5b7a9562f4d
|
3 |
size 939771079
|
pytorch_model-00005-of-00014.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 939796331
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:87cdf5be9be6a8d991928ed6732c5cc90375859df14159775b950c980db41dbd
|
3 |
size 939796331
|
pytorch_model-00006-of-00014.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 939778659
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d47eeeccf7caa4cdd9085d3a1dac9009bf4585bd42efb5b3880ab5ed627b30ed
|
3 |
size 939778659
|
pytorch_model-00007-of-00014.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 939771079
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:201f81cf5d72ea37104c1a8418363af76ed491c050498d00cdb823b03f1cb48e
|
3 |
size 939771079
|
pytorch_model-00008-of-00014.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 939796331
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b130dfdb83723317822917aecf7b6fd0d41237e2ae3f5d8c19be40d180a04a86
|
3 |
size 939796331
|
pytorch_model-00009-of-00014.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 939778659
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:adbfe7bed837484fadeb5b9b94275c648df0bf5e7b27a9188f0ca2920e9b77b1
|
3 |
size 939778659
|
pytorch_model-00010-of-00014.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 939771079
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:896a77fa5cf15d206e6e682aafb1d49f2498042a8b12aa3abe209f44a6ded08d
|
3 |
size 939771079
|
pytorch_model-00011-of-00014.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 939796331
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:539c16613ece36a91c1c8ae384c72537b9b1f155a165822168dcb3a7448b6444
|
3 |
size 939796331
|
pytorch_model-00012-of-00014.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 939778659
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c84f0e1a5e55169d1165b7f39020025471fae0b355a24d249d311d14b7089edf
|
3 |
size 939778659
|
pytorch_model-00013-of-00014.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 939771079
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:97d2e16c7b07fa8820c92c79262c5ea8f9724244bb3ef5216d25114f3a5d3b53
|
3 |
size 939771079
|
pytorch_model-00014-of-00014.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 927052345
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:50dc17a8d73316eabd8f6d3b0dc7be5efe5db648313c3c26365915a9513d87e9
|
3 |
size 927052345
|
special_tokens_map.json
CHANGED
@@ -1,30 +1,11 @@
|
|
1 |
{
|
2 |
-
"
|
3 |
-
"
|
4 |
-
"
|
5 |
-
"
|
6 |
-
|
7 |
-
|
8 |
-
|
9 |
-
"
|
10 |
-
|
11 |
-
"lstrip": false,
|
12 |
-
"normalized": true,
|
13 |
-
"rstrip": false,
|
14 |
-
"single_word": false
|
15 |
-
},
|
16 |
-
"pad_token": {
|
17 |
-
"content": "<unk>",
|
18 |
-
"lstrip": false,
|
19 |
-
"normalized": true,
|
20 |
-
"rstrip": false,
|
21 |
-
"single_word": false
|
22 |
-
},
|
23 |
-
"unk_token": {
|
24 |
-
"content": "<unk>",
|
25 |
-
"lstrip": false,
|
26 |
-
"normalized": true,
|
27 |
-
"rstrip": false,
|
28 |
-
"single_word": false
|
29 |
-
}
|
30 |
}
|
|
|
1 |
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"<unk>",
|
4 |
+
"<s>",
|
5 |
+
"</s>"
|
6 |
+
],
|
7 |
+
"bos_token": "<s>",
|
8 |
+
"eos_token": "</s>",
|
9 |
+
"pad_token": "<unk>",
|
10 |
+
"unk_token": "<unk>"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
}
|
tokenizer.json
CHANGED
@@ -9,7 +9,7 @@
|
|
9 |
"single_word": false,
|
10 |
"lstrip": false,
|
11 |
"rstrip": false,
|
12 |
-
"normalized":
|
13 |
"special": true
|
14 |
},
|
15 |
{
|
@@ -18,7 +18,7 @@
|
|
18 |
"single_word": false,
|
19 |
"lstrip": false,
|
20 |
"rstrip": false,
|
21 |
-
"normalized":
|
22 |
"special": true
|
23 |
},
|
24 |
{
|
@@ -27,7 +27,7 @@
|
|
27 |
"single_word": false,
|
28 |
"lstrip": false,
|
29 |
"rstrip": false,
|
30 |
-
"normalized":
|
31 |
"special": true
|
32 |
}
|
33 |
],
|
@@ -51,6 +51,12 @@
|
|
51 |
"post_processor": {
|
52 |
"type": "TemplateProcessing",
|
53 |
"single": [
|
|
|
|
|
|
|
|
|
|
|
|
|
54 |
{
|
55 |
"Sequence": {
|
56 |
"id": "A",
|
@@ -59,12 +65,24 @@
|
|
59 |
}
|
60 |
],
|
61 |
"pair": [
|
|
|
|
|
|
|
|
|
|
|
|
|
62 |
{
|
63 |
"Sequence": {
|
64 |
"id": "A",
|
65 |
"type_id": 0
|
66 |
}
|
67 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
68 |
{
|
69 |
"Sequence": {
|
70 |
"id": "B",
|
@@ -72,7 +90,17 @@
|
|
72 |
}
|
73 |
}
|
74 |
],
|
75 |
-
"special_tokens": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
76 |
},
|
77 |
"decoder": {
|
78 |
"type": "Sequence",
|
|
|
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 |
],
|
|
|
51 |
"post_processor": {
|
52 |
"type": "TemplateProcessing",
|
53 |
"single": [
|
54 |
+
{
|
55 |
+
"SpecialToken": {
|
56 |
+
"id": "<s>",
|
57 |
+
"type_id": 0
|
58 |
+
}
|
59 |
+
},
|
60 |
{
|
61 |
"Sequence": {
|
62 |
"id": "A",
|
|
|
65 |
}
|
66 |
],
|
67 |
"pair": [
|
68 |
+
{
|
69 |
+
"SpecialToken": {
|
70 |
+
"id": "<s>",
|
71 |
+
"type_id": 0
|
72 |
+
}
|
73 |
+
},
|
74 |
{
|
75 |
"Sequence": {
|
76 |
"id": "A",
|
77 |
"type_id": 0
|
78 |
}
|
79 |
},
|
80 |
+
{
|
81 |
+
"SpecialToken": {
|
82 |
+
"id": "<s>",
|
83 |
+
"type_id": 1
|
84 |
+
}
|
85 |
+
},
|
86 |
{
|
87 |
"Sequence": {
|
88 |
"id": "B",
|
|
|
90 |
}
|
91 |
}
|
92 |
],
|
93 |
+
"special_tokens": {
|
94 |
+
"<s>": {
|
95 |
+
"id": "<s>",
|
96 |
+
"ids": [
|
97 |
+
1
|
98 |
+
],
|
99 |
+
"tokens": [
|
100 |
+
"<s>"
|
101 |
+
]
|
102 |
+
}
|
103 |
+
}
|
104 |
},
|
105 |
"decoder": {
|
106 |
"type": "Sequence",
|
tokenizer_config.json
CHANGED
@@ -1,42 +1,44 @@
|
|
1 |
{
|
2 |
-
"
|
3 |
-
|
4 |
-
|
5 |
-
|
6 |
-
|
7 |
-
|
8 |
-
|
9 |
-
|
10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
"clean_up_tokenization_spaces": false,
|
13 |
-
"eos_token":
|
14 |
-
"__type": "AddedToken",
|
15 |
-
"content": "</s>",
|
16 |
-
"lstrip": false,
|
17 |
-
"normalized": true,
|
18 |
-
"rstrip": false,
|
19 |
-
"single_word": false
|
20 |
-
},
|
21 |
"legacy": true,
|
22 |
"model_max_length": 1000000000000000019884624838656,
|
23 |
-
"pad_token":
|
24 |
-
"__type": "AddedToken",
|
25 |
-
"content": "<unk>",
|
26 |
-
"lstrip": false,
|
27 |
-
"normalized": true,
|
28 |
-
"rstrip": false,
|
29 |
-
"single_word": false
|
30 |
-
},
|
31 |
"padding_side": "left",
|
32 |
"sp_model_kwargs": {},
|
33 |
"tokenizer_class": "LlamaTokenizer",
|
34 |
-
"unk_token":
|
35 |
-
|
36 |
-
"content": "<unk>",
|
37 |
-
"lstrip": false,
|
38 |
-
"normalized": true,
|
39 |
-
"rstrip": false,
|
40 |
-
"single_word": false
|
41 |
-
}
|
42 |
}
|
|
|
1 |
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "<unk>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "<s>",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "</s>",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
}
|
27 |
},
|
28 |
+
"additional_special_tokens": [
|
29 |
+
"<unk>",
|
30 |
+
"<s>",
|
31 |
+
"</s>"
|
32 |
+
],
|
33 |
+
"bos_token": "<s>",
|
34 |
"clean_up_tokenization_spaces": false,
|
35 |
+
"eos_token": "</s>",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
36 |
"legacy": true,
|
37 |
"model_max_length": 1000000000000000019884624838656,
|
38 |
+
"pad_token": "<unk>",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
39 |
"padding_side": "left",
|
40 |
"sp_model_kwargs": {},
|
41 |
"tokenizer_class": "LlamaTokenizer",
|
42 |
+
"unk_token": "<unk>",
|
43 |
+
"use_default_system_prompt": true
|
|
|
|
|
|
|
|
|
|
|
|
|
44 |
}
|