Upload 3 files
Browse files- added_tokens.json +98 -0
- config.json +80 -0
- gitattributes +19 -0
added_tokens.json
ADDED
@@ -0,0 +1,98 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<|dummy_0|>": 100256,
|
3 |
+
"<|endoftext|>": 100257,
|
4 |
+
"<|fim_prefix|>": 100258,
|
5 |
+
"<|fim_middle|>": 100259,
|
6 |
+
"<|fim_suffix|>": 100260,
|
7 |
+
"<|dummy_1|>": 100261,
|
8 |
+
"<|dummy_2|>": 100262,
|
9 |
+
"<|dummy_3|>": 100263,
|
10 |
+
"<|im_start|>": 100264,
|
11 |
+
"<|im_end|>": 100265,
|
12 |
+
"<|im_sep|>": 100266,
|
13 |
+
"<|dummy_4|>": 100267,
|
14 |
+
"<|dummy_5|>": 100268,
|
15 |
+
"<|dummy_6|>": 100269,
|
16 |
+
"<|dummy_7|>": 100270,
|
17 |
+
"<|dummy_8|>": 100271,
|
18 |
+
"<|dummy_9|>": 100272,
|
19 |
+
"<|dummy_10|>": 100273,
|
20 |
+
"<|dummy_11|>": 100274,
|
21 |
+
"<|dummy_12|>": 100275,
|
22 |
+
"<|endofprompt|>": 100276,
|
23 |
+
"<|dummy_13|>": 100277,
|
24 |
+
"<|dummy_14|>": 100278,
|
25 |
+
"<|dummy_15|>": 100279,
|
26 |
+
"<|dummy_16|>": 100280,
|
27 |
+
"<|dummy_17|>": 100281,
|
28 |
+
"<|dummy_18|>": 100282,
|
29 |
+
"<|dummy_19|>": 100283,
|
30 |
+
"<|dummy_20|>": 100284,
|
31 |
+
"<|dummy_21|>": 100285,
|
32 |
+
"<|dummy_22|>": 100286,
|
33 |
+
"<|dummy_23|>": 100287,
|
34 |
+
"<|dummy_24|>": 100288,
|
35 |
+
"<|dummy_25|>": 100289,
|
36 |
+
"<|dummy_26|>": 100290,
|
37 |
+
"<|dummy_27|>": 100291,
|
38 |
+
"<|dummy_28|>": 100292,
|
39 |
+
"<|dummy_29|>": 100293,
|
40 |
+
"<|dummy_30|>": 100294,
|
41 |
+
"<|dummy_31|>": 100295,
|
42 |
+
"<|dummy_32|>": 100296,
|
43 |
+
"<|dummy_33|>": 100297,
|
44 |
+
"<|dummy_34|>": 100298,
|
45 |
+
"<|dummy_35|>": 100299,
|
46 |
+
"<|dummy_36|>": 100300,
|
47 |
+
"<|dummy_37|>": 100301,
|
48 |
+
"<|dummy_38|>": 100302,
|
49 |
+
"<|dummy_39|>": 100303,
|
50 |
+
"<|dummy_40|>": 100304,
|
51 |
+
"<|dummy_41|>": 100305,
|
52 |
+
"<|dummy_42|>": 100306,
|
53 |
+
"<|dummy_43|>": 100307,
|
54 |
+
"<|dummy_44|>": 100308,
|
55 |
+
"<|dummy_45|>": 100309,
|
56 |
+
"<|dummy_46|>": 100310,
|
57 |
+
"<|dummy_47|>": 100311,
|
58 |
+
"<|dummy_48|>": 100312,
|
59 |
+
"<|dummy_49|>": 100313,
|
60 |
+
"<|dummy_50|>": 100314,
|
61 |
+
"<|dummy_51|>": 100315,
|
62 |
+
"<|dummy_52|>": 100316,
|
63 |
+
"<|dummy_53|>": 100317,
|
64 |
+
"<|dummy_54|>": 100318,
|
65 |
+
"<|dummy_55|>": 100319,
|
66 |
+
"<|dummy_56|>": 100320,
|
67 |
+
"<|dummy_57|>": 100321,
|
68 |
+
"<|dummy_58|>": 100322,
|
69 |
+
"<|dummy_59|>": 100323,
|
70 |
+
"<|dummy_60|>": 100324,
|
71 |
+
"<|dummy_61|>": 100325,
|
72 |
+
"<|dummy_62|>": 100326,
|
73 |
+
"<|dummy_63|>": 100327,
|
74 |
+
"<|dummy_64|>": 100328,
|
75 |
+
"<|dummy_65|>": 100329,
|
76 |
+
"<|dummy_66|>": 100330,
|
77 |
+
"<|dummy_67|>": 100331,
|
78 |
+
"<|dummy_68|>": 100332,
|
79 |
+
"<|dummy_69|>": 100333,
|
80 |
+
"<|dummy_70|>": 100334,
|
81 |
+
"<|dummy_71|>": 100335,
|
82 |
+
"<|dummy_72|>": 100336,
|
83 |
+
"<|dummy_73|>": 100337,
|
84 |
+
"<|dummy_74|>": 100338,
|
85 |
+
"<|dummy_75|>": 100339,
|
86 |
+
"<|dummy_76|>": 100340,
|
87 |
+
"<|dummy_77|>": 100341,
|
88 |
+
"<|dummy_78|>": 100342,
|
89 |
+
"<|dummy_79|>": 100343,
|
90 |
+
"<|dummy_80|>": 100344,
|
91 |
+
"<|dummy_81|>": 100345,
|
92 |
+
"<|dummy_82|>": 100346,
|
93 |
+
"<|dummy_83|>": 100347,
|
94 |
+
"<|dummy_84|>": 100348,
|
95 |
+
"<|dummy_85|>": 100349,
|
96 |
+
"<|dummy_86|>": 100350,
|
97 |
+
"<|dummy_87|>": 100351
|
98 |
+
}
|
config.json
ADDED
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"activation_function": "gelu_new",
|
3 |
+
"architectures": [
|
4 |
+
"GPTNeoForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_dropout": 0,
|
7 |
+
"attention_layers": [
|
8 |
+
"global",
|
9 |
+
"local",
|
10 |
+
"global",
|
11 |
+
"local",
|
12 |
+
"global",
|
13 |
+
"local",
|
14 |
+
"global",
|
15 |
+
"local",
|
16 |
+
"global",
|
17 |
+
"local",
|
18 |
+
"global",
|
19 |
+
"local",
|
20 |
+
"global",
|
21 |
+
"local",
|
22 |
+
"global",
|
23 |
+
"local",
|
24 |
+
"global",
|
25 |
+
"local",
|
26 |
+
"global",
|
27 |
+
"local",
|
28 |
+
"global",
|
29 |
+
"local",
|
30 |
+
"global",
|
31 |
+
"local",
|
32 |
+
"global",
|
33 |
+
"local",
|
34 |
+
"global",
|
35 |
+
"local",
|
36 |
+
"global",
|
37 |
+
"local",
|
38 |
+
"global",
|
39 |
+
"local"
|
40 |
+
],
|
41 |
+
"attention_types": [
|
42 |
+
[
|
43 |
+
[
|
44 |
+
"global",
|
45 |
+
"local"
|
46 |
+
],
|
47 |
+
16
|
48 |
+
]
|
49 |
+
],
|
50 |
+
"bos_token_id": 50256,
|
51 |
+
"embed_dropout": 0,
|
52 |
+
"eos_token_id": 50256,
|
53 |
+
"gradient_checkpointing": false,
|
54 |
+
"hidden_size": 2560,
|
55 |
+
"initializer_range": 0.02,
|
56 |
+
"intermediate_size": null,
|
57 |
+
"layer_norm_epsilon": 1e-05,
|
58 |
+
"max_position_embeddings": 2048,
|
59 |
+
"model_type": "gpt_neo",
|
60 |
+
"num_heads": 20,
|
61 |
+
"num_layers": 32,
|
62 |
+
"resid_dropout": 0,
|
63 |
+
"summary_activation": null,
|
64 |
+
"summary_first_dropout": 0.1,
|
65 |
+
"summary_proj_to_labels": true,
|
66 |
+
"summary_type": "cls_index",
|
67 |
+
"summary_use_proj": true,
|
68 |
+
"task_specific_params": {
|
69 |
+
"text-generation": {
|
70 |
+
"do_sample": true,
|
71 |
+
"max_length": 50,
|
72 |
+
"temperature": 0.9
|
73 |
+
}
|
74 |
+
},
|
75 |
+
"tokenizer_class": "GPT2Tokenizer",
|
76 |
+
"transformers_version": "4.9.0.dev0",
|
77 |
+
"use_cache": true,
|
78 |
+
"vocab_size": 50257,
|
79 |
+
"window_size": 256
|
80 |
+
}
|
gitattributes
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
*.bin.* filter=lfs diff=lfs merge=lfs -text
|
2 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
4 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
5 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
6 |
+
*.tar.gz filter=lfs diff=lfs merge=lfs -text
|
7 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
8 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
9 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
10 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
11 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
14 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
15 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
16 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
17 |
+
pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
|
18 |
+
rust_model.ot filter=lfs diff=lfs merge=lfs -text
|
19 |
+
model.safetensors filter=lfs diff=lfs merge=lfs -text
|