Hugging Face's logo

gowitheflowlab
/
clip-wiki-pretrain-2560-22800

PyTorch
clip
Model card Files Community
gowitheflow1998 commited on
Commit
a337d56
·
1 Parent(s): d0c3f07

upload checkpoint

Browse files
Files changed (16) hide show
  1. .gitattributes +5 -0
  2. 49e6dc219d1a1a1c9236acaf05a48b542002016a6dc877ee72baab085a84257b.3f28e7f4b38e1efe1b6da4a3732404c19d4c6a614ff32dce90a251e293d4ce58 +3 -0
  3. config.json +166 -0
  4. merges.txt +0 -0
  5. optimizer.pt +3 -0
  6. preprocessor_config.json +19 -0
  7. pytorch_model.bin +3 -0
  8. rng_state.pth +3 -0
  9. scheduler.pt +3 -0
  10. special_tokens_map.json +1 -0
  11. text_renderer_config.json +253 -0
  12. tokenizer.json +0 -0
  13. tokenizer_config.json +34 -0
  14. trainer_state.json +1156 -0
  15. training_args.bin +3 -0
  16. vocab.json +0 -0
.gitattributes CHANGED
@@ -33,3 +33,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
37
+ training_args.bin filter=lfs diff=lfs merge=lfs -text
38
+ optimizer.pt filter=lfs diff=lfs merge=lfs -text
39
+ scheduler.pt filter=lfs diff=lfs merge=lfs -text
40
+ 49e6dc219d1a1a1c9236acaf05a48b542002016a6dc877ee72baab085a84257b.3f28e7f4b38e1efe1b6da4a3732404c19d4c6a614ff32dce90a251e293d4ce58 filter=lfs diff=lfs merge=lfs -text
49e6dc219d1a1a1c9236acaf05a48b542002016a6dc877ee72baab085a84257b.3f28e7f4b38e1efe1b6da4a3732404c19d4c6a614ff32dce90a251e293d4ce58 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83ab5c39e2b1c34a955136275ce0db068cb20d9643ead033d6b8124a73ab4f64
3
+ size 15645492
config.json ADDED
@@ -0,0 +1,166 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openai/clip-vit-base-patch16",
3
+ "architectures": [
4
+ "CLIPModel"
5
+ ],
6
+ "id2label": {},
7
+ "initializer_factor": 1.0,
8
+ "label2id": {
9
+ "Contradiction": 2,
10
+ "Entailment": 0,
11
+ "Neutral": 1
12
+ },
13
+ "logit_scale_init_value": 2.6592,
14
+ "model_type": "clip",
15
+ "projection_dim": 512,
16
+ "text_config": {
17
+ "_name_or_path": "",
18
+ "add_cross_attention": false,
19
+ "architectures": null,
20
+ "attention_dropout": 0.0,
21
+ "bad_words_ids": null,
22
+ "bos_token_id": 0,
23
+ "chunk_size_feed_forward": 0,
24
+ "cross_attention_hidden_size": null,
25
+ "decoder_start_token_id": null,
26
+ "diversity_penalty": 0.0,
27
+ "do_sample": false,
28
+ "dropout": 0.0,
29
+ "early_stopping": false,
30
+ "encoder_no_repeat_ngram_size": 0,
31
+ "eos_token_id": 2,
32
+ "finetuning_task": null,
33
+ "forced_bos_token_id": null,
34
+ "forced_eos_token_id": null,
35
+ "hidden_act": "quick_gelu",
36
+ "hidden_size": 512,
37
+ "id2label": {
38
+ "0": "LABEL_0",
39
+ "1": "LABEL_1"
40
+ },
41
+ "initializer_factor": 1.0,
42
+ "initializer_range": 0.02,
43
+ "intermediate_size": 2048,
44
+ "is_decoder": false,
45
+ "is_encoder_decoder": false,
46
+ "label2id": {
47
+ "LABEL_0": 0,
48
+ "LABEL_1": 1
49
+ },
50
+ "layer_norm_eps": 1e-05,
51
+ "length_penalty": 1.0,
52
+ "max_length": 20,
53
+ "max_position_embeddings": 77,
54
+ "min_length": 0,
55
+ "model_type": "clip_text_model",
56
+ "no_repeat_ngram_size": 0,
57
+ "num_attention_heads": 8,
58
+ "num_beam_groups": 1,
59
+ "num_beams": 1,
60
+ "num_hidden_layers": 12,
61
+ "num_return_sequences": 1,
62
+ "output_attentions": false,
63
+ "output_hidden_states": false,
64
+ "output_scores": false,
65
+ "pad_token_id": 1,
66
+ "prefix": null,
67
+ "problem_type": null,
68
+ "pruned_heads": {},
69
+ "remove_invalid_values": false,
70
+ "repetition_penalty": 1.0,
71
+ "return_dict": true,
72
+ "return_dict_in_generate": false,
73
+ "sep_token_id": null,
74
+ "task_specific_params": null,
75
+ "temperature": 1.0,
76
+ "tie_encoder_decoder": false,
77
+ "tie_word_embeddings": true,
78
+ "tokenizer_class": null,
79
+ "top_k": 50,
80
+ "top_p": 1.0,
81
+ "torch_dtype": null,
82
+ "torchscript": false,
83
+ "transformers_version": "4.17.0",
84
+ "typical_p": 1.0,
85
+ "use_bfloat16": false,
86
+ "vocab_size": 49408
87
+ },
88
+ "text_config_dict": null,
89
+ "torch_dtype": "float32",
90
+ "transformers_version": null,
91
+ "vision_config": {
92
+ "_name_or_path": "",
93
+ "add_cross_attention": false,
94
+ "architectures": null,
95
+ "attention_dropout": 0.0,
96
+ "bad_words_ids": null,
97
+ "bos_token_id": null,
98
+ "chunk_size_feed_forward": 0,
99
+ "cross_attention_hidden_size": null,
100
+ "decoder_start_token_id": null,
101
+ "diversity_penalty": 0.0,
102
+ "do_sample": false,
103
+ "dropout": 0.0,
104
+ "early_stopping": false,
105
+ "encoder_no_repeat_ngram_size": 0,
106
+ "eos_token_id": null,
107
+ "finetuning_task": null,
108
+ "forced_bos_token_id": null,
109
+ "forced_eos_token_id": null,
110
+ "hidden_act": "quick_gelu",
111
+ "hidden_size": 768,
112
+ "id2label": {
113
+ "0": "LABEL_0",
114
+ "1": "LABEL_1"
115
+ },
116
+ "image_size": 224,
117
+ "initializer_factor": 1.0,
118
+ "initializer_range": 0.02,
119
+ "intermediate_size": 3072,
120
+ "is_decoder": false,
121
+ "is_encoder_decoder": false,
122
+ "label2id": {
123
+ "LABEL_0": 0,
124
+ "LABEL_1": 1
125
+ },
126
+ "layer_norm_eps": 1e-05,
127
+ "length_penalty": 1.0,
128
+ "max_length": 20,
129
+ "min_length": 0,
130
+ "model_type": "clip_vision_model",
131
+ "no_repeat_ngram_size": 0,
132
+ "num_attention_heads": 12,
133
+ "num_beam_groups": 1,
134
+ "num_beams": 1,
135
+ "num_hidden_layers": 12,
136
+ "num_return_sequences": 1,
137
+ "output_attentions": false,
138
+ "output_hidden_states": false,
139
+ "output_scores": false,
140
+ "pad_token_id": null,
141
+ "patch_size": 16,
142
+ "prefix": null,
143
+ "problem_type": null,
144
+ "pruned_heads": {},
145
+ "remove_invalid_values": false,
146
+ "repetition_penalty": 1.0,
147
+ "return_dict": true,
148
+ "return_dict_in_generate": false,
149
+ "sep_token_id": null,
150
+ "task_specific_params": null,
151
+ "temperature": 1.0,
152
+ "tie_encoder_decoder": false,
153
+ "tie_word_embeddings": true,
154
+ "tokenizer_class": null,
155
+ "top_k": 50,
156
+ "top_p": 1.0,
157
+ "torch_dtype": null,
158
+ "torchscript": false,
159
+ "transformers_version": "4.17.0",
160
+ "typical_p": 1.0,
161
+ "use_bfloat16": false
162
+ },
163
+ "vision_config_dict": {
164
+ "patch_size": 16
165
+ }
166
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d70f1bc20b56950e9fd3bf0831441d9055e59a08a61b15588e889bb16b73fde2
3
+ size 689659578
preprocessor_config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": 224,
3
+ "do_center_crop": true,
4
+ "do_normalize": true,
5
+ "do_resize": true,
6
+ "feature_extractor_type": "CLIPFeatureExtractor",
7
+ "image_mean": [
8
+ 0.48145466,
9
+ 0.4578275,
10
+ 0.40821073
11
+ ],
12
+ "image_std": [
13
+ 0.26862954,
14
+ 0.26130258,
15
+ 0.27577711
16
+ ],
17
+ "resample": 3,
18
+ "size": 224
19
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c25bf4ef34bd3b8b204aa20767bfa67eb17be98309238623fea19aa7f694455
3
+ size 598642122
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93b830fe93e6f7710b58e80e3a93850a14d38db75d558fb5feca0a9d01caf620
3
+ size 14244
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8166e01d2b10247de29befc9577caa94ee2add44c473badc8c77fad131eccddd
3
+ size 1064
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": {"content": "<|startoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "eos_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "unk_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "pad_token": "<|endoftext|>"}
text_renderer_config.json ADDED
@@ -0,0 +1,253 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "background_color": "white",
3
+ "dpi": 120,
4
+ "font_color": "black",
5
+ "font_file": "49e6dc219d1a1a1c9236acaf05a48b542002016a6dc877ee72baab085a84257b.3f28e7f4b38e1efe1b6da4a3732404c19d4c6a614ff32dce90a251e293d4ce58",
6
+ "font_size": 8,
7
+ "fonts_list": [
8
+ "Apple Color Emoji",
9
+ "DejaVu Sans",
10
+ "DejaVu Sans Mono",
11
+ "DejaVu Serif",
12
+ "FontAwesome",
13
+ "GLYPHICONS Halflings",
14
+ "Go Noto Current",
15
+ "Inconsolata",
16
+ "Lato",
17
+ "MathJax_AMS",
18
+ "MathJax_Caligraphic",
19
+ "MathJax_Fraktur",
20
+ "MathJax_Main",
21
+ "MathJax_Math",
22
+ "MathJax_SansSerif",
23
+ "MathJax_Script",
24
+ "MathJax_Size1",
25
+ "MathJax_Size2",
26
+ "MathJax_Size3",
27
+ "MathJax_Size4",
28
+ "MathJax_Typewriter",
29
+ "MathJax_Vector",
30
+ "MathJax_Vector-Bold",
31
+ "MathJax_WinChrome",
32
+ "MathJax_WinIE6",
33
+ "Monospace",
34
+ "Noto Fangsong KSS Rotated",
35
+ "Noto Fangsong KSS Vertical",
36
+ "Noto Kufi Arabic",
37
+ "Noto Music",
38
+ "Noto Naskh Arabic UI",
39
+ "Noto Nastaliq Urdu",
40
+ "Noto Rashi Hebrew",
41
+ "Noto Sans",
42
+ "Noto Sans Adlam",
43
+ "Noto Sans Adlam Unjoined",
44
+ "Noto Sans Anatolian Hieroglyphs",
45
+ "Noto Sans Arabic",
46
+ "Noto Sans Armenian",
47
+ "Noto Sans Avestan",
48
+ "Noto Sans Balinese",
49
+ "Noto Sans Bamum",
50
+ "Noto Sans Bassa Vah",
51
+ "Noto Sans Batak",
52
+ "Noto Sans Bhaiksuki",
53
+ "Noto Sans Brahmi",
54
+ "Noto Sans Buginese",
55
+ "Noto Sans Buhid",
56
+ "Noto Sans CJK HK",
57
+ "Noto Sans CJK JP",
58
+ "Noto Sans CJK KR",
59
+ "Noto Sans CJK SC",
60
+ "Noto Sans CJK TC",
61
+ "Noto Sans Canadian Aboriginal",
62
+ "Noto Sans Carian",
63
+ "Noto Sans Caucasian Albanian",
64
+ "Noto Sans Chakma",
65
+ "Noto Sans Cham",
66
+ "Noto Sans Cherokee",
67
+ "Noto Sans Chorasmian",
68
+ "Noto Sans Coptic",
69
+ "Noto Sans Cuneiform",
70
+ "Noto Sans Cypriot",
71
+ "Noto Sans Cypro Minoan",
72
+ "Noto Sans Deseret",
73
+ "Noto Sans Devanagari",
74
+ "Noto Sans Devanagari UI",
75
+ "Noto Sans Duployan",
76
+ "Noto Sans Egyptian Hieroglyphs",
77
+ "Noto Sans Elbasan",
78
+ "Noto Sans Elymaic",
79
+ "Noto Sans Ethiopic",
80
+ "Noto Sans Georgian",
81
+ "Noto Sans Glagolitic",
82
+ "Noto Sans Gothic",
83
+ "Noto Sans Grantha",
84
+ "Noto Sans Gujarati",
85
+ "Noto Sans Gujarati UI",
86
+ "Noto Sans Gunjala Gondi",
87
+ "Noto Sans Gurmukhi",
88
+ "Noto Sans Gurmukhi UI",
89
+ "Noto Sans Hanifi Rohingya",
90
+ "Noto Sans Hanunoo",
91
+ "Noto Sans Hatran",
92
+ "Noto Sans Hebrew",
93
+ "Noto Sans Hebrew Droid",
94
+ "Noto Sans Imperial Aramaic",
95
+ "Noto Sans Indic Siyaq Numbers",
96
+ "Noto Sans Inscriptional Pahlavi",
97
+ "Noto Sans Inscriptional Parthian",
98
+ "Noto Sans Javanese",
99
+ "Noto Sans Kaithi",
100
+ "Noto Sans Kannada",
101
+ "Noto Sans Kannada UI",
102
+ "Noto Sans Kawi",
103
+ "Noto Sans Kayah Li",
104
+ "Noto Sans Kharoshthi",
105
+ "Noto Sans Khmer",
106
+ "Noto Sans Khmer UI",
107
+ "Noto Sans Khojki",
108
+ "Noto Sans Khudawadi",
109
+ "Noto Sans Lao",
110
+ "Noto Sans Lao Looped",
111
+ "Noto Sans Lao Looped UI",
112
+ "Noto Sans Lao UI",
113
+ "Noto Sans Lepcha",
114
+ "Noto Sans Limbu",
115
+ "Noto Sans Linear A",
116
+ "Noto Sans Linear B",
117
+ "Noto Sans Lisu",
118
+ "Noto Sans Lycian",
119
+ "Noto Sans Lydian",
120
+ "Noto Sans Mahajani",
121
+ "Noto Sans Malayalam",
122
+ "Noto Sans Malayalam UI",
123
+ "Noto Sans Mandaic",
124
+ "Noto Sans Manichaean",
125
+ "Noto Sans Marchen",
126
+ "Noto Sans Masaram Gondi",
127
+ "Noto Sans Math",
128
+ "Noto Sans Mayan Numerals",
129
+ "Noto Sans Medefaidrin",
130
+ "Noto Sans Meetei Mayek",
131
+ "Noto Sans Mende Kikakui",
132
+ "Noto Sans Meroitic",
133
+ "Noto Sans Miao",
134
+ "Noto Sans Modi",
135
+ "Noto Sans Mongolian",
136
+ "Noto Sans Mono",
137
+ "Noto Sans Mro",
138
+ "Noto Sans Multani",
139
+ "Noto Sans Myanmar",
140
+ "Noto Sans Myanmar UI",
141
+ "Noto Sans NKo",
142
+ "Noto Sans NKo Unjoined",
143
+ "Noto Sans Nabataean",
144
+ "Noto Sans Nag Mundari",
145
+ "Noto Sans Nandinagari",
146
+ "Noto Sans New Tai Lue",
147
+ "Noto Sans Newa",
148
+ "Noto Sans Nushu",
149
+ "Noto Sans Ogham",
150
+ "Noto Sans Ol Chiki",
151
+ "Noto Sans Old Hungarian",
152
+ "Noto Sans Old Italic",
153
+ "Noto Sans Old North Arabian",
154
+ "Noto Sans Old Permic",
155
+ "Noto Sans Old Persian",
156
+ "Noto Sans Old Sogdian",
157
+ "Noto Sans Old South Arabian",
158
+ "Noto Sans Old Turkic",
159
+ "Noto Sans Oriya",
160
+ "Noto Sans Osage",
161
+ "Noto Sans Osmanya",
162
+ "Noto Sans Pahawh Hmong",
163
+ "Noto Sans Palmyrene",
164
+ "Noto Sans Pau Cin Hau",
165
+ "Noto Sans PhagsPa",
166
+ "Noto Sans Phoenician",
167
+ "Noto Sans Psalter Pahlavi",
168
+ "Noto Sans Rejang",
169
+ "Noto Sans Runic",
170
+ "Noto Sans Samaritan",
171
+ "Noto Sans Saurashtra",
172
+ "Noto Sans Sharada",
173
+ "Noto Sans Shavian",
174
+ "Noto Sans Siddham",
175
+ "Noto Sans SignWriting",
176
+ "Noto Sans Sinhala",
177
+ "Noto Sans Sinhala UI",
178
+ "Noto Sans Sogdian",
179
+ "Noto Sans Sora Sompeng",
180
+ "Noto Sans Soyombo",
181
+ "Noto Sans Sundanese",
182
+ "Noto Sans Syloti Nagri",
183
+ "Noto Sans Symbols",
184
+ "Noto Sans Symbols 2",
185
+ "Noto Sans Syriac",
186
+ "Noto Sans Syriac Eastern",
187
+ "Noto Sans Syriac Western",
188
+ "Noto Sans Tagalog",
189
+ "Noto Sans Tagbanwa",
190
+ "Noto Sans Tai Le",
191
+ "Noto Sans Tai Tham",
192
+ "Noto Sans Tai Viet",
193
+ "Noto Sans Takri",
194
+ "Noto Sans Tamil",
195
+ "Noto Sans Tamil Supplement",
196
+ "Noto Sans Tamil UI",
197
+ "Noto Sans Tangsa",
198
+ "Noto Sans Telugu",
199
+ "Noto Sans Telugu UI",
200
+ "Noto Sans Test",
201
+ "Noto Sans Thaana",
202
+ "Noto Sans Thai",
203
+ "Noto Sans Thai Looped",
204
+ "Noto Sans Thai Looped UI",
205
+ "Noto Sans Thai UI",
206
+ "Noto Sans Tifinagh",
207
+ "Noto Sans Tifinagh APT",
208
+ "Noto Sans Tifinagh Adrar",
209
+ "Noto Sans Tifinagh Agraw Imazighen",
210
+ "Noto Sans Tifinagh Ahaggar",
211
+ "Noto Sans Tifinagh Air",
212
+ "Noto Sans Tifinagh Azawagh",
213
+ "Noto Sans Tifinagh Ghat",
214
+ "Noto Sans Tifinagh Hawad",
215
+ "Noto Sans Tifinagh Rhissa Ixa",
216
+ "Noto Sans Tifinagh SIL",
217
+ "Noto Sans Tifinagh Tawellemmet",
218
+ "Noto Sans Tirhuta",
219
+ "Noto Sans Ugaritic",
220
+ "Noto Sans Vai",
221
+ "Noto Sans Vithkuqi",
222
+ "Noto Sans Wancho",
223
+ "Noto Sans Warang Citi",
224
+ "Noto Sans Yi",
225
+ "Noto Sans Zanabazar Square",
226
+ "Noto Traditional Nushu",
227
+ "NotoSansOldHungarianUI",
228
+ "Sans",
229
+ "Serif",
230
+ "Source Code Pro",
231
+ "System-ui",
232
+ "Ubuntu",
233
+ "Ubuntu Condensed",
234
+ "Ubuntu Mono",
235
+ "cmex10",
236
+ "cmmi10",
237
+ "cmr10",
238
+ "cmsy10",
239
+ "dsrom10",
240
+ "esint10",
241
+ "eufm10",
242
+ "msam10",
243
+ "msbm10",
244
+ "rsfs10",
245
+ "stmary10",
246
+ "wasy10"
247
+ ],
248
+ "max_seq_length": 196,
249
+ "pad_size": 3,
250
+ "pixels_per_patch": 16,
251
+ "rgb": false,
252
+ "text_renderer_type": "PangoCairoTextRenderer"
253
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "unk_token": {
3
+ "content": "<|endoftext|>",
4
+ "single_word": false,
5
+ "lstrip": false,
6
+ "rstrip": false,
7
+ "normalized": true,
8
+ "__type": "AddedToken"
9
+ },
10
+ "bos_token": {
11
+ "content": "<|startoftext|>",
12
+ "single_word": false,
13
+ "lstrip": false,
14
+ "rstrip": false,
15
+ "normalized": true,
16
+ "__type": "AddedToken"
17
+ },
18
+ "eos_token": {
19
+ "content": "<|endoftext|>",
20
+ "single_word": false,
21
+ "lstrip": false,
22
+ "rstrip": false,
23
+ "normalized": true,
24
+ "__type": "AddedToken"
25
+ },
26
+ "pad_token": "<|endoftext|>",
27
+ "errors": "replace",
28
+ "add_prefix_space": false,
29
+ "do_lower_case": true,
30
+ "name_or_path": "openai/clip-vit-base-patch16",
31
+ "model_max_length": 77,
32
+ "special_tokens_map_file": "./special_tokens_map.json",
33
+ "tokenizer_class": "CLIPTokenizer"
34
+ }
trainer_state.json ADDED
@@ -0,0 +1,1156 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7275755258839849,
3
+ "best_model_checkpoint": "./checkpoints/clip-2560-wikispan-all/checkpoint-240",
4
+ "epoch": 0.9472765798329802,
5
+ "global_step": 22800,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.0,
12
+ "learning_rate": 1.8e-06,
13
+ "loss": 1.8723,
14
+ "step": 120
15
+ },
16
+ {
17
+ "epoch": 0.01,
18
+ "learning_rate": 3.6e-06,
19
+ "loss": 1.0969,
20
+ "step": 240
21
+ },
22
+ {
23
+ "epoch": 0.01,
24
+ "learning_rate": 5.4e-06,
25
+ "loss": 0.8982,
26
+ "step": 360
27
+ },
28
+ {
29
+ "epoch": 0.02,
30
+ "learning_rate": 7.2e-06,
31
+ "loss": 0.802,
32
+ "step": 480
33
+ },
34
+ {
35
+ "epoch": 0.02,
36
+ "learning_rate": 9e-06,
37
+ "loss": 0.7304,
38
+ "step": 600
39
+ },
40
+ {
41
+ "epoch": 0.03,
42
+ "learning_rate": 1.08e-05,
43
+ "loss": 0.6815,
44
+ "step": 720
45
+ },
46
+ {
47
+ "epoch": 0.03,
48
+ "learning_rate": 1.26e-05,
49
+ "loss": 0.6435,
50
+ "step": 840
51
+ },
52
+ {
53
+ "epoch": 0.04,
54
+ "learning_rate": 1.44e-05,
55
+ "loss": 0.6089,
56
+ "step": 960
57
+ },
58
+ {
59
+ "epoch": 0.04,
60
+ "learning_rate": 1.62e-05,
61
+ "loss": 0.581,
62
+ "step": 1080
63
+ },
64
+ {
65
+ "epoch": 0.05,
66
+ "learning_rate": 1.8e-05,
67
+ "loss": 0.5637,
68
+ "step": 1200
69
+ },
70
+ {
71
+ "epoch": 0.05,
72
+ "learning_rate": 1.98e-05,
73
+ "loss": 0.538,
74
+ "step": 1320
75
+ },
76
+ {
77
+ "epoch": 0.06,
78
+ "learning_rate": 2.16e-05,
79
+ "loss": 0.5223,
80
+ "step": 1440
81
+ },
82
+ {
83
+ "epoch": 0.06,
84
+ "learning_rate": 2.3400000000000003e-05,
85
+ "loss": 0.5123,
86
+ "step": 1560
87
+ },
88
+ {
89
+ "epoch": 0.07,
90
+ "learning_rate": 2.52e-05,
91
+ "loss": 0.4984,
92
+ "step": 1680
93
+ },
94
+ {
95
+ "epoch": 0.07,
96
+ "learning_rate": 2.7000000000000002e-05,
97
+ "loss": 0.4848,
98
+ "step": 1800
99
+ },
100
+ {
101
+ "epoch": 0.08,
102
+ "learning_rate": 2.88e-05,
103
+ "loss": 0.4763,
104
+ "step": 1920
105
+ },
106
+ {
107
+ "epoch": 0.08,
108
+ "learning_rate": 2.9945750452079567e-05,
109
+ "loss": 0.468,
110
+ "step": 2040
111
+ },
112
+ {
113
+ "epoch": 0.09,
114
+ "learning_rate": 2.978300180831826e-05,
115
+ "loss": 0.4532,
116
+ "step": 2160
117
+ },
118
+ {
119
+ "epoch": 0.09,
120
+ "learning_rate": 2.9620253164556963e-05,
121
+ "loss": 0.445,
122
+ "step": 2280
123
+ },
124
+ {
125
+ "epoch": 0.1,
126
+ "learning_rate": 2.945750452079566e-05,
127
+ "loss": 0.4382,
128
+ "step": 2400
129
+ },
130
+ {
131
+ "epoch": 0.1,
132
+ "learning_rate": 2.929475587703436e-05,
133
+ "loss": 0.4289,
134
+ "step": 2520
135
+ },
136
+ {
137
+ "epoch": 0.11,
138
+ "learning_rate": 2.9132007233273057e-05,
139
+ "loss": 0.4216,
140
+ "step": 2640
141
+ },
142
+ {
143
+ "epoch": 0.11,
144
+ "learning_rate": 2.8969258589511756e-05,
145
+ "loss": 0.4134,
146
+ "step": 2760
147
+ },
148
+ {
149
+ "epoch": 0.12,
150
+ "learning_rate": 2.8806509945750454e-05,
151
+ "loss": 0.4084,
152
+ "step": 2880
153
+ },
154
+ {
155
+ "epoch": 0.12,
156
+ "learning_rate": 2.864376130198915e-05,
157
+ "loss": 0.3988,
158
+ "step": 3000
159
+ },
160
+ {
161
+ "epoch": 0.13,
162
+ "learning_rate": 2.8481012658227846e-05,
163
+ "loss": 0.3938,
164
+ "step": 3120
165
+ },
166
+ {
167
+ "epoch": 0.13,
168
+ "learning_rate": 2.8318264014466548e-05,
169
+ "loss": 0.3886,
170
+ "step": 3240
171
+ },
172
+ {
173
+ "epoch": 0.14,
174
+ "learning_rate": 2.8155515370705246e-05,
175
+ "loss": 0.3845,
176
+ "step": 3360
177
+ },
178
+ {
179
+ "epoch": 0.14,
180
+ "learning_rate": 2.7992766726943944e-05,
181
+ "loss": 0.378,
182
+ "step": 3480
183
+ },
184
+ {
185
+ "epoch": 0.15,
186
+ "learning_rate": 2.7830018083182642e-05,
187
+ "loss": 0.375,
188
+ "step": 3600
189
+ },
190
+ {
191
+ "epoch": 0.15,
192
+ "learning_rate": 2.766726943942134e-05,
193
+ "loss": 0.3665,
194
+ "step": 3720
195
+ },
196
+ {
197
+ "epoch": 0.16,
198
+ "learning_rate": 2.7504520795660035e-05,
199
+ "loss": 0.3644,
200
+ "step": 3840
201
+ },
202
+ {
203
+ "epoch": 0.16,
204
+ "learning_rate": 2.7341772151898733e-05,
205
+ "loss": 0.36,
206
+ "step": 3960
207
+ },
208
+ {
209
+ "epoch": 0.17,
210
+ "learning_rate": 2.717902350813743e-05,
211
+ "loss": 0.3597,
212
+ "step": 4080
213
+ },
214
+ {
215
+ "epoch": 0.17,
216
+ "learning_rate": 2.701627486437613e-05,
217
+ "loss": 0.3549,
218
+ "step": 4200
219
+ },
220
+ {
221
+ "epoch": 0.18,
222
+ "learning_rate": 2.685352622061483e-05,
223
+ "loss": 0.3545,
224
+ "step": 4320
225
+ },
226
+ {
227
+ "epoch": 0.18,
228
+ "learning_rate": 2.669077757685353e-05,
229
+ "loss": 0.3496,
230
+ "step": 4440
231
+ },
232
+ {
233
+ "epoch": 0.19,
234
+ "learning_rate": 2.6528028933092224e-05,
235
+ "loss": 0.3442,
236
+ "step": 4560
237
+ },
238
+ {
239
+ "epoch": 0.19,
240
+ "learning_rate": 2.6365280289330922e-05,
241
+ "loss": 0.3424,
242
+ "step": 4680
243
+ },
244
+ {
245
+ "epoch": 0.2,
246
+ "learning_rate": 2.620253164556962e-05,
247
+ "loss": 0.3414,
248
+ "step": 4800
249
+ },
250
+ {
251
+ "epoch": 0.2,
252
+ "learning_rate": 2.6039783001808318e-05,
253
+ "loss": 0.3361,
254
+ "step": 4920
255
+ },
256
+ {
257
+ "epoch": 0.21,
258
+ "learning_rate": 2.5877034358047016e-05,
259
+ "loss": 0.3378,
260
+ "step": 5040
261
+ },
262
+ {
263
+ "epoch": 0.21,
264
+ "learning_rate": 2.5714285714285714e-05,
265
+ "loss": 0.3316,
266
+ "step": 5160
267
+ },
268
+ {
269
+ "epoch": 0.22,
270
+ "learning_rate": 2.5551537070524416e-05,
271
+ "loss": 0.3302,
272
+ "step": 5280
273
+ },
274
+ {
275
+ "epoch": 0.22,
276
+ "learning_rate": 2.538878842676311e-05,
277
+ "loss": 0.3245,
278
+ "step": 5400
279
+ },
280
+ {
281
+ "epoch": 0.23,
282
+ "learning_rate": 2.522603978300181e-05,
283
+ "loss": 0.3254,
284
+ "step": 5520
285
+ },
286
+ {
287
+ "epoch": 0.23,
288
+ "learning_rate": 2.5063291139240507e-05,
289
+ "loss": 0.3206,
290
+ "step": 5640
291
+ },
292
+ {
293
+ "epoch": 0.24,
294
+ "learning_rate": 2.4900542495479205e-05,
295
+ "loss": 0.3212,
296
+ "step": 5760
297
+ },
298
+ {
299
+ "epoch": 0.24,
300
+ "learning_rate": 2.4737793851717903e-05,
301
+ "loss": 0.3184,
302
+ "step": 5880
303
+ },
304
+ {
305
+ "epoch": 0.25,
306
+ "learning_rate": 2.45750452079566e-05,
307
+ "loss": 0.3141,
308
+ "step": 6000
309
+ },
310
+ {
311
+ "epoch": 0.25,
312
+ "learning_rate": 2.44122965641953e-05,
313
+ "loss": 0.3185,
314
+ "step": 6120
315
+ },
316
+ {
317
+ "epoch": 0.26,
318
+ "learning_rate": 2.4249547920433994e-05,
319
+ "loss": 0.3115,
320
+ "step": 6240
321
+ },
322
+ {
323
+ "epoch": 0.26,
324
+ "learning_rate": 2.4086799276672696e-05,
325
+ "loss": 0.3093,
326
+ "step": 6360
327
+ },
328
+ {
329
+ "epoch": 0.27,
330
+ "learning_rate": 2.3924050632911394e-05,
331
+ "loss": 0.3116,
332
+ "step": 6480
333
+ },
334
+ {
335
+ "epoch": 0.27,
336
+ "learning_rate": 2.3761301989150092e-05,
337
+ "loss": 0.3084,
338
+ "step": 6600
339
+ },
340
+ {
341
+ "epoch": 0.28,
342
+ "learning_rate": 2.359855334538879e-05,
343
+ "loss": 0.3071,
344
+ "step": 6720
345
+ },
346
+ {
347
+ "epoch": 0.28,
348
+ "learning_rate": 2.3435804701627488e-05,
349
+ "loss": 0.3047,
350
+ "step": 6840
351
+ },
352
+ {
353
+ "epoch": 0.29,
354
+ "learning_rate": 2.3273056057866183e-05,
355
+ "loss": 0.302,
356
+ "step": 6960
357
+ },
358
+ {
359
+ "epoch": 0.29,
360
+ "learning_rate": 2.311030741410488e-05,
361
+ "loss": 0.3027,
362
+ "step": 7080
363
+ },
364
+ {
365
+ "epoch": 0.3,
366
+ "learning_rate": 2.294755877034358e-05,
367
+ "loss": 0.3019,
368
+ "step": 7200
369
+ },
370
+ {
371
+ "epoch": 0.3,
372
+ "learning_rate": 2.278481012658228e-05,
373
+ "loss": 0.3,
374
+ "step": 7320
375
+ },
376
+ {
377
+ "epoch": 0.31,
378
+ "learning_rate": 2.262206148282098e-05,
379
+ "loss": 0.2961,
380
+ "step": 7440
381
+ },
382
+ {
383
+ "epoch": 0.31,
384
+ "learning_rate": 2.2459312839059677e-05,
385
+ "loss": 0.2967,
386
+ "step": 7560
387
+ },
388
+ {
389
+ "epoch": 0.32,
390
+ "learning_rate": 2.2296564195298375e-05,
391
+ "loss": 0.2944,
392
+ "step": 7680
393
+ },
394
+ {
395
+ "epoch": 0.32,
396
+ "learning_rate": 2.213381555153707e-05,
397
+ "loss": 0.2937,
398
+ "step": 7800
399
+ },
400
+ {
401
+ "epoch": 0.33,
402
+ "learning_rate": 2.1971066907775768e-05,
403
+ "loss": 0.2914,
404
+ "step": 7920
405
+ },
406
+ {
407
+ "epoch": 0.33,
408
+ "learning_rate": 2.1808318264014466e-05,
409
+ "loss": 0.2902,
410
+ "step": 8040
411
+ },
412
+ {
413
+ "epoch": 0.34,
414
+ "learning_rate": 2.1645569620253164e-05,
415
+ "loss": 0.2889,
416
+ "step": 8160
417
+ },
418
+ {
419
+ "epoch": 0.34,
420
+ "learning_rate": 2.1482820976491862e-05,
421
+ "loss": 0.2894,
422
+ "step": 8280
423
+ },
424
+ {
425
+ "epoch": 0.35,
426
+ "learning_rate": 2.1320072332730564e-05,
427
+ "loss": 0.286,
428
+ "step": 8400
429
+ },
430
+ {
431
+ "epoch": 0.35,
432
+ "learning_rate": 2.1157323688969262e-05,
433
+ "loss": 0.2847,
434
+ "step": 8520
435
+ },
436
+ {
437
+ "epoch": 0.36,
438
+ "learning_rate": 2.0994575045207956e-05,
439
+ "loss": 0.2838,
440
+ "step": 8640
441
+ },
442
+ {
443
+ "epoch": 0.36,
444
+ "learning_rate": 2.0831826401446655e-05,
445
+ "loss": 0.2822,
446
+ "step": 8760
447
+ },
448
+ {
449
+ "epoch": 0.37,
450
+ "learning_rate": 2.0669077757685353e-05,
451
+ "loss": 0.2814,
452
+ "step": 8880
453
+ },
454
+ {
455
+ "epoch": 0.37,
456
+ "learning_rate": 2.050632911392405e-05,
457
+ "loss": 0.2818,
458
+ "step": 9000
459
+ },
460
+ {
461
+ "epoch": 0.38,
462
+ "learning_rate": 2.034358047016275e-05,
463
+ "loss": 0.2779,
464
+ "step": 9120
465
+ },
466
+ {
467
+ "epoch": 0.38,
468
+ "learning_rate": 2.0180831826401447e-05,
469
+ "loss": 0.2794,
470
+ "step": 9240
471
+ },
472
+ {
473
+ "epoch": 0.39,
474
+ "learning_rate": 2.0018083182640145e-05,
475
+ "loss": 0.2789,
476
+ "step": 9360
477
+ },
478
+ {
479
+ "epoch": 0.39,
480
+ "learning_rate": 1.9855334538878843e-05,
481
+ "loss": 0.2757,
482
+ "step": 9480
483
+ },
484
+ {
485
+ "epoch": 0.4,
486
+ "learning_rate": 1.969258589511754e-05,
487
+ "loss": 0.2758,
488
+ "step": 9600
489
+ },
490
+ {
491
+ "epoch": 0.4,
492
+ "learning_rate": 1.952983725135624e-05,
493
+ "loss": 0.277,
494
+ "step": 9720
495
+ },
496
+ {
497
+ "epoch": 0.41,
498
+ "learning_rate": 1.9367088607594938e-05,
499
+ "loss": 0.2731,
500
+ "step": 9840
501
+ },
502
+ {
503
+ "epoch": 0.41,
504
+ "learning_rate": 1.9204339963833636e-05,
505
+ "loss": 0.268,
506
+ "step": 9960
507
+ },
508
+ {
509
+ "epoch": 0.42,
510
+ "learning_rate": 1.9041591320072334e-05,
511
+ "loss": 0.2702,
512
+ "step": 10080
513
+ },
514
+ {
515
+ "epoch": 0.42,
516
+ "learning_rate": 1.887884267631103e-05,
517
+ "loss": 0.2699,
518
+ "step": 10200
519
+ },
520
+ {
521
+ "epoch": 0.43,
522
+ "learning_rate": 1.8716094032549727e-05,
523
+ "loss": 0.2707,
524
+ "step": 10320
525
+ },
526
+ {
527
+ "epoch": 0.43,
528
+ "learning_rate": 1.8553345388788428e-05,
529
+ "loss": 0.2661,
530
+ "step": 10440
531
+ },
532
+ {
533
+ "epoch": 0.44,
534
+ "learning_rate": 1.8390596745027126e-05,
535
+ "loss": 0.2668,
536
+ "step": 10560
537
+ },
538
+ {
539
+ "epoch": 0.44,
540
+ "learning_rate": 1.8227848101265824e-05,
541
+ "loss": 0.2689,
542
+ "step": 10680
543
+ },
544
+ {
545
+ "epoch": 0.45,
546
+ "learning_rate": 1.8065099457504523e-05,
547
+ "loss": 0.2695,
548
+ "step": 10800
549
+ },
550
+ {
551
+ "epoch": 0.45,
552
+ "learning_rate": 1.790235081374322e-05,
553
+ "loss": 0.2637,
554
+ "step": 10920
555
+ },
556
+ {
557
+ "epoch": 0.46,
558
+ "learning_rate": 1.7739602169981915e-05,
559
+ "loss": 0.2645,
560
+ "step": 11040
561
+ },
562
+ {
563
+ "epoch": 0.46,
564
+ "learning_rate": 1.7576853526220614e-05,
565
+ "loss": 0.2618,
566
+ "step": 11160
567
+ },
568
+ {
569
+ "epoch": 0.47,
570
+ "learning_rate": 1.741410488245931e-05,
571
+ "loss": 0.2626,
572
+ "step": 11280
573
+ },
574
+ {
575
+ "epoch": 0.47,
576
+ "learning_rate": 1.7251356238698013e-05,
577
+ "loss": 0.2581,
578
+ "step": 11400
579
+ },
580
+ {
581
+ "epoch": 0.48,
582
+ "learning_rate": 1.708860759493671e-05,
583
+ "loss": 0.2597,
584
+ "step": 11520
585
+ },
586
+ {
587
+ "epoch": 0.48,
588
+ "learning_rate": 1.692585895117541e-05,
589
+ "loss": 0.2594,
590
+ "step": 11640
591
+ },
592
+ {
593
+ "epoch": 0.49,
594
+ "learning_rate": 1.6763110307414104e-05,
595
+ "loss": 0.2577,
596
+ "step": 11760
597
+ },
598
+ {
599
+ "epoch": 0.49,
600
+ "learning_rate": 1.6600361663652802e-05,
601
+ "loss": 0.2585,
602
+ "step": 11880
603
+ },
604
+ {
605
+ "epoch": 0.5,
606
+ "learning_rate": 1.64376130198915e-05,
607
+ "loss": 0.2565,
608
+ "step": 12000
609
+ },
610
+ {
611
+ "epoch": 0.5,
612
+ "learning_rate": 1.62748643761302e-05,
613
+ "loss": 0.2562,
614
+ "step": 12120
615
+ },
616
+ {
617
+ "epoch": 0.51,
618
+ "learning_rate": 1.6112115732368897e-05,
619
+ "loss": 0.2559,
620
+ "step": 12240
621
+ },
622
+ {
623
+ "epoch": 0.51,
624
+ "learning_rate": 1.5949367088607595e-05,
625
+ "loss": 0.2551,
626
+ "step": 12360
627
+ },
628
+ {
629
+ "epoch": 0.52,
630
+ "learning_rate": 1.5786618444846296e-05,
631
+ "loss": 0.2528,
632
+ "step": 12480
633
+ },
634
+ {
635
+ "epoch": 0.52,
636
+ "learning_rate": 1.562386980108499e-05,
637
+ "loss": 0.2539,
638
+ "step": 12600
639
+ },
640
+ {
641
+ "epoch": 0.53,
642
+ "learning_rate": 1.546112115732369e-05,
643
+ "loss": 0.2525,
644
+ "step": 12720
645
+ },
646
+ {
647
+ "epoch": 0.53,
648
+ "learning_rate": 1.5298372513562387e-05,
649
+ "loss": 0.2534,
650
+ "step": 12840
651
+ },
652
+ {
653
+ "epoch": 0.54,
654
+ "learning_rate": 1.5135623869801085e-05,
655
+ "loss": 0.2518,
656
+ "step": 12960
657
+ },
658
+ {
659
+ "epoch": 0.54,
660
+ "learning_rate": 1.4972875226039783e-05,
661
+ "loss": 0.2506,
662
+ "step": 13080
663
+ },
664
+ {
665
+ "epoch": 0.55,
666
+ "learning_rate": 1.4810126582278482e-05,
667
+ "loss": 0.2517,
668
+ "step": 13200
669
+ },
670
+ {
671
+ "epoch": 0.55,
672
+ "learning_rate": 1.464737793851718e-05,
673
+ "loss": 0.2483,
674
+ "step": 13320
675
+ },
676
+ {
677
+ "epoch": 0.56,
678
+ "learning_rate": 1.4484629294755878e-05,
679
+ "loss": 0.2471,
680
+ "step": 13440
681
+ },
682
+ {
683
+ "epoch": 0.56,
684
+ "learning_rate": 1.4321880650994574e-05,
685
+ "loss": 0.2484,
686
+ "step": 13560
687
+ },
688
+ {
689
+ "epoch": 0.57,
690
+ "learning_rate": 1.4159132007233274e-05,
691
+ "loss": 0.2474,
692
+ "step": 13680
693
+ },
694
+ {
695
+ "epoch": 0.57,
696
+ "learning_rate": 1.3996383363471972e-05,
697
+ "loss": 0.25,
698
+ "step": 13800
699
+ },
700
+ {
701
+ "epoch": 0.58,
702
+ "learning_rate": 1.383363471971067e-05,
703
+ "loss": 0.2458,
704
+ "step": 13920
705
+ },
706
+ {
707
+ "epoch": 0.58,
708
+ "learning_rate": 1.3670886075949367e-05,
709
+ "loss": 0.2466,
710
+ "step": 14040
711
+ },
712
+ {
713
+ "epoch": 0.59,
714
+ "learning_rate": 1.3508137432188065e-05,
715
+ "loss": 0.2454,
716
+ "step": 14160
717
+ },
718
+ {
719
+ "epoch": 0.59,
720
+ "learning_rate": 1.3345388788426765e-05,
721
+ "loss": 0.2461,
722
+ "step": 14280
723
+ },
724
+ {
725
+ "epoch": 0.6,
726
+ "learning_rate": 1.3182640144665461e-05,
727
+ "loss": 0.2427,
728
+ "step": 14400
729
+ },
730
+ {
731
+ "epoch": 0.6,
732
+ "learning_rate": 1.3019891500904159e-05,
733
+ "loss": 0.2446,
734
+ "step": 14520
735
+ },
736
+ {
737
+ "epoch": 0.61,
738
+ "learning_rate": 1.2857142857142857e-05,
739
+ "loss": 0.2433,
740
+ "step": 14640
741
+ },
742
+ {
743
+ "epoch": 0.61,
744
+ "learning_rate": 1.2694394213381555e-05,
745
+ "loss": 0.241,
746
+ "step": 14760
747
+ },
748
+ {
749
+ "epoch": 0.62,
750
+ "learning_rate": 1.2531645569620253e-05,
751
+ "loss": 0.2424,
752
+ "step": 14880
753
+ },
754
+ {
755
+ "epoch": 0.62,
756
+ "learning_rate": 1.2368896925858952e-05,
757
+ "loss": 0.2427,
758
+ "step": 15000
759
+ },
760
+ {
761
+ "epoch": 0.63,
762
+ "learning_rate": 1.220614828209765e-05,
763
+ "loss": 0.2395,
764
+ "step": 15120
765
+ },
766
+ {
767
+ "epoch": 0.63,
768
+ "learning_rate": 1.2043399638336348e-05,
769
+ "loss": 0.2396,
770
+ "step": 15240
771
+ },
772
+ {
773
+ "epoch": 0.64,
774
+ "learning_rate": 1.1880650994575046e-05,
775
+ "loss": 0.2416,
776
+ "step": 15360
777
+ },
778
+ {
779
+ "epoch": 0.64,
780
+ "learning_rate": 1.1717902350813744e-05,
781
+ "loss": 0.2402,
782
+ "step": 15480
783
+ },
784
+ {
785
+ "epoch": 0.65,
786
+ "learning_rate": 1.155515370705244e-05,
787
+ "loss": 0.2382,
788
+ "step": 15600
789
+ },
790
+ {
791
+ "epoch": 0.65,
792
+ "learning_rate": 1.139240506329114e-05,
793
+ "loss": 0.2361,
794
+ "step": 15720
795
+ },
796
+ {
797
+ "epoch": 0.66,
798
+ "learning_rate": 1.1229656419529838e-05,
799
+ "loss": 0.2383,
800
+ "step": 15840
801
+ },
802
+ {
803
+ "epoch": 0.66,
804
+ "learning_rate": 1.1066907775768535e-05,
805
+ "loss": 0.2368,
806
+ "step": 15960
807
+ },
808
+ {
809
+ "epoch": 0.67,
810
+ "learning_rate": 1.0904159132007233e-05,
811
+ "loss": 0.2374,
812
+ "step": 16080
813
+ },
814
+ {
815
+ "epoch": 0.67,
816
+ "learning_rate": 1.0741410488245931e-05,
817
+ "loss": 0.2363,
818
+ "step": 16200
819
+ },
820
+ {
821
+ "epoch": 0.68,
822
+ "learning_rate": 1.0578661844484631e-05,
823
+ "loss": 0.2364,
824
+ "step": 16320
825
+ },
826
+ {
827
+ "epoch": 0.68,
828
+ "learning_rate": 1.0415913200723327e-05,
829
+ "loss": 0.2326,
830
+ "step": 16440
831
+ },
832
+ {
833
+ "epoch": 0.69,
834
+ "learning_rate": 1.0253164556962025e-05,
835
+ "loss": 0.232,
836
+ "step": 16560
837
+ },
838
+ {
839
+ "epoch": 0.69,
840
+ "learning_rate": 1.0090415913200724e-05,
841
+ "loss": 0.2317,
842
+ "step": 16680
843
+ },
844
+ {
845
+ "epoch": 0.7,
846
+ "learning_rate": 9.927667269439422e-06,
847
+ "loss": 0.2316,
848
+ "step": 16800
849
+ },
850
+ {
851
+ "epoch": 0.7,
852
+ "learning_rate": 9.76491862567812e-06,
853
+ "loss": 0.2343,
854
+ "step": 16920
855
+ },
856
+ {
857
+ "epoch": 0.71,
858
+ "learning_rate": 9.602169981916818e-06,
859
+ "loss": 0.2304,
860
+ "step": 17040
861
+ },
862
+ {
863
+ "epoch": 0.71,
864
+ "learning_rate": 9.439421338155514e-06,
865
+ "loss": 0.2297,
866
+ "step": 17160
867
+ },
868
+ {
869
+ "epoch": 0.72,
870
+ "learning_rate": 9.276672694394214e-06,
871
+ "loss": 0.231,
872
+ "step": 17280
873
+ },
874
+ {
875
+ "epoch": 0.72,
876
+ "learning_rate": 9.113924050632912e-06,
877
+ "loss": 0.2316,
878
+ "step": 17400
879
+ },
880
+ {
881
+ "epoch": 0.73,
882
+ "learning_rate": 8.95117540687161e-06,
883
+ "loss": 0.2314,
884
+ "step": 17520
885
+ },
886
+ {
887
+ "epoch": 0.73,
888
+ "learning_rate": 8.788426763110307e-06,
889
+ "loss": 0.2289,
890
+ "step": 17640
891
+ },
892
+ {
893
+ "epoch": 0.74,
894
+ "learning_rate": 8.625678119349007e-06,
895
+ "loss": 0.2275,
896
+ "step": 17760
897
+ },
898
+ {
899
+ "epoch": 0.74,
900
+ "learning_rate": 8.462929475587705e-06,
901
+ "loss": 0.2282,
902
+ "step": 17880
903
+ },
904
+ {
905
+ "epoch": 0.75,
906
+ "learning_rate": 8.300180831826401e-06,
907
+ "loss": 0.2285,
908
+ "step": 18000
909
+ },
910
+ {
911
+ "epoch": 0.75,
912
+ "learning_rate": 8.1374321880651e-06,
913
+ "loss": 0.2284,
914
+ "step": 18120
915
+ },
916
+ {
917
+ "epoch": 0.76,
918
+ "learning_rate": 7.974683544303797e-06,
919
+ "loss": 0.2274,
920
+ "step": 18240
921
+ },
922
+ {
923
+ "epoch": 0.76,
924
+ "learning_rate": 7.811934900542495e-06,
925
+ "loss": 0.2234,
926
+ "step": 18360
927
+ },
928
+ {
929
+ "epoch": 0.77,
930
+ "learning_rate": 7.649186256781194e-06,
931
+ "loss": 0.229,
932
+ "step": 18480
933
+ },
934
+ {
935
+ "epoch": 0.77,
936
+ "learning_rate": 7.486437613019892e-06,
937
+ "loss": 0.2248,
938
+ "step": 18600
939
+ },
940
+ {
941
+ "epoch": 0.78,
942
+ "learning_rate": 7.32368896925859e-06,
943
+ "loss": 0.2232,
944
+ "step": 18720
945
+ },
946
+ {
947
+ "epoch": 0.78,
948
+ "learning_rate": 7.160940325497287e-06,
949
+ "loss": 0.2234,
950
+ "step": 18840
951
+ },
952
+ {
953
+ "epoch": 0.79,
954
+ "learning_rate": 6.998191681735986e-06,
955
+ "loss": 0.2258,
956
+ "step": 18960
957
+ },
958
+ {
959
+ "epoch": 0.79,
960
+ "learning_rate": 6.835443037974683e-06,
961
+ "loss": 0.2238,
962
+ "step": 19080
963
+ },
964
+ {
965
+ "epoch": 0.8,
966
+ "learning_rate": 6.672694394213382e-06,
967
+ "loss": 0.2217,
968
+ "step": 19200
969
+ },
970
+ {
971
+ "epoch": 0.8,
972
+ "learning_rate": 6.5099457504520796e-06,
973
+ "loss": 0.2222,
974
+ "step": 19320
975
+ },
976
+ {
977
+ "epoch": 0.81,
978
+ "learning_rate": 6.347197106690778e-06,
979
+ "loss": 0.2256,
980
+ "step": 19440
981
+ },
982
+ {
983
+ "epoch": 0.81,
984
+ "learning_rate": 6.184448462929476e-06,
985
+ "loss": 0.2221,
986
+ "step": 19560
987
+ },
988
+ {
989
+ "epoch": 0.82,
990
+ "learning_rate": 6.021699819168174e-06,
991
+ "loss": 0.2243,
992
+ "step": 19680
993
+ },
994
+ {
995
+ "epoch": 0.82,
996
+ "learning_rate": 5.858951175406872e-06,
997
+ "loss": 0.2227,
998
+ "step": 19800
999
+ },
1000
+ {
1001
+ "epoch": 0.83,
1002
+ "learning_rate": 5.69620253164557e-06,
1003
+ "loss": 0.222,
1004
+ "step": 19920
1005
+ },
1006
+ {
1007
+ "epoch": 0.83,
1008
+ "learning_rate": 5.533453887884267e-06,
1009
+ "loss": 0.2205,
1010
+ "step": 20040
1011
+ },
1012
+ {
1013
+ "epoch": 0.84,
1014
+ "learning_rate": 5.3707052441229655e-06,
1015
+ "loss": 0.22,
1016
+ "step": 20160
1017
+ },
1018
+ {
1019
+ "epoch": 0.84,
1020
+ "learning_rate": 5.207956600361664e-06,
1021
+ "loss": 0.2223,
1022
+ "step": 20280
1023
+ },
1024
+ {
1025
+ "epoch": 0.85,
1026
+ "learning_rate": 5.045207956600362e-06,
1027
+ "loss": 0.2209,
1028
+ "step": 20400
1029
+ },
1030
+ {
1031
+ "epoch": 0.85,
1032
+ "learning_rate": 4.88245931283906e-06,
1033
+ "loss": 0.2207,
1034
+ "step": 20520
1035
+ },
1036
+ {
1037
+ "epoch": 0.86,
1038
+ "learning_rate": 4.719710669077757e-06,
1039
+ "loss": 0.2206,
1040
+ "step": 20640
1041
+ },
1042
+ {
1043
+ "epoch": 0.86,
1044
+ "learning_rate": 4.556962025316456e-06,
1045
+ "loss": 0.2204,
1046
+ "step": 20760
1047
+ },
1048
+ {
1049
+ "epoch": 0.87,
1050
+ "learning_rate": 4.394213381555153e-06,
1051
+ "loss": 0.2221,
1052
+ "step": 20880
1053
+ },
1054
+ {
1055
+ "epoch": 0.87,
1056
+ "learning_rate": 4.231464737793852e-06,
1057
+ "loss": 0.2197,
1058
+ "step": 21000
1059
+ },
1060
+ {
1061
+ "epoch": 0.88,
1062
+ "learning_rate": 4.06871609403255e-06,
1063
+ "loss": 0.2188,
1064
+ "step": 21120
1065
+ },
1066
+ {
1067
+ "epoch": 0.88,
1068
+ "learning_rate": 3.905967450271248e-06,
1069
+ "loss": 0.2202,
1070
+ "step": 21240
1071
+ },
1072
+ {
1073
+ "epoch": 0.89,
1074
+ "learning_rate": 3.743218806509946e-06,
1075
+ "loss": 0.2176,
1076
+ "step": 21360
1077
+ },
1078
+ {
1079
+ "epoch": 0.89,
1080
+ "learning_rate": 3.5804701627486435e-06,
1081
+ "loss": 0.2185,
1082
+ "step": 21480
1083
+ },
1084
+ {
1085
+ "epoch": 0.9,
1086
+ "learning_rate": 3.4177215189873417e-06,
1087
+ "loss": 0.2169,
1088
+ "step": 21600
1089
+ },
1090
+ {
1091
+ "epoch": 0.9,
1092
+ "learning_rate": 3.2549728752260398e-06,
1093
+ "loss": 0.2175,
1094
+ "step": 21720
1095
+ },
1096
+ {
1097
+ "epoch": 0.91,
1098
+ "learning_rate": 3.092224231464738e-06,
1099
+ "loss": 0.2153,
1100
+ "step": 21840
1101
+ },
1102
+ {
1103
+ "epoch": 0.91,
1104
+ "learning_rate": 2.929475587703436e-06,
1105
+ "loss": 0.218,
1106
+ "step": 21960
1107
+ },
1108
+ {
1109
+ "epoch": 0.92,
1110
+ "learning_rate": 2.7667269439421337e-06,
1111
+ "loss": 0.2161,
1112
+ "step": 22080
1113
+ },
1114
+ {
1115
+ "epoch": 0.92,
1116
+ "learning_rate": 2.603978300180832e-06,
1117
+ "loss": 0.2167,
1118
+ "step": 22200
1119
+ },
1120
+ {
1121
+ "epoch": 0.93,
1122
+ "learning_rate": 2.44122965641953e-06,
1123
+ "loss": 0.2166,
1124
+ "step": 22320
1125
+ },
1126
+ {
1127
+ "epoch": 0.93,
1128
+ "learning_rate": 2.278481012658228e-06,
1129
+ "loss": 0.215,
1130
+ "step": 22440
1131
+ },
1132
+ {
1133
+ "epoch": 0.94,
1134
+ "learning_rate": 2.115732368896926e-06,
1135
+ "loss": 0.2145,
1136
+ "step": 22560
1137
+ },
1138
+ {
1139
+ "epoch": 0.94,
1140
+ "learning_rate": 1.952983725135624e-06,
1141
+ "loss": 0.2165,
1142
+ "step": 22680
1143
+ },
1144
+ {
1145
+ "epoch": 0.95,
1146
+ "learning_rate": 1.7902350813743218e-06,
1147
+ "loss": 0.2155,
1148
+ "step": 22800
1149
+ }
1150
+ ],
1151
+ "max_steps": 24120,
1152
+ "num_train_epochs": 2,
1153
+ "total_flos": 0.0,
1154
+ "trial_name": null,
1155
+ "trial_params": null
1156
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b4e9b908ba7cfd25eb41cec551f2d6f2b9d63b77ff4fceb67a9c378ae462e18
3
+ size 3640
vocab.json ADDED
The diff for this file is too large to render. See raw diff