mrm8488 commited on
Commit
dda18f1
1 Parent(s): 991014e

Initial commit from mrm8488

Browse files
config.json ADDED
@@ -0,0 +1,168 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/content/drive/MyDrive/r2r_shared_mlsum-fr-3/checkpoint-60000",
3
+ "architectures": [
4
+ "EncoderDecoderModel"
5
+ ],
6
+ "decoder": {
7
+ "_name_or_path": "camembert-base",
8
+ "add_cross_attention": true,
9
+ "architectures": [
10
+ "CamembertForMaskedLM"
11
+ ],
12
+ "attention_probs_dropout_prob": 0.1,
13
+ "bad_words_ids": null,
14
+ "bos_token_id": 5,
15
+ "chunk_size_feed_forward": 0,
16
+ "decoder_start_token_id": null,
17
+ "diversity_penalty": 0.0,
18
+ "do_sample": false,
19
+ "early_stopping": false,
20
+ "encoder_no_repeat_ngram_size": 0,
21
+ "eos_token_id": 6,
22
+ "finetuning_task": null,
23
+ "forced_bos_token_id": null,
24
+ "forced_eos_token_id": null,
25
+ "gradient_checkpointing": false,
26
+ "hidden_act": "gelu",
27
+ "hidden_dropout_prob": 0.1,
28
+ "hidden_size": 768,
29
+ "id2label": {
30
+ "0": "LABEL_0",
31
+ "1": "LABEL_1"
32
+ },
33
+ "initializer_range": 0.02,
34
+ "intermediate_size": 3072,
35
+ "is_decoder": true,
36
+ "is_encoder_decoder": false,
37
+ "label2id": {
38
+ "LABEL_0": 0,
39
+ "LABEL_1": 1
40
+ },
41
+ "layer_norm_eps": 1e-05,
42
+ "length_penalty": 1.0,
43
+ "max_length": 20,
44
+ "max_position_embeddings": 514,
45
+ "min_length": 0,
46
+ "model_type": "camembert",
47
+ "no_repeat_ngram_size": 0,
48
+ "num_attention_heads": 12,
49
+ "num_beam_groups": 1,
50
+ "num_beams": 1,
51
+ "num_hidden_layers": 12,
52
+ "num_return_sequences": 1,
53
+ "output_attentions": false,
54
+ "output_hidden_states": false,
55
+ "output_past": true,
56
+ "output_scores": false,
57
+ "pad_token_id": 1,
58
+ "position_embedding_type": "absolute",
59
+ "prefix": null,
60
+ "problem_type": null,
61
+ "pruned_heads": {},
62
+ "remove_invalid_values": false,
63
+ "repetition_penalty": 1.0,
64
+ "return_dict": true,
65
+ "return_dict_in_generate": false,
66
+ "sep_token_id": null,
67
+ "task_specific_params": null,
68
+ "temperature": 1.0,
69
+ "tie_encoder_decoder": false,
70
+ "tie_word_embeddings": true,
71
+ "tokenizer_class": null,
72
+ "top_k": 50,
73
+ "top_p": 1.0,
74
+ "torchscript": false,
75
+ "transformers_version": "4.7.0.dev0",
76
+ "type_vocab_size": 1,
77
+ "use_bfloat16": false,
78
+ "use_cache": true,
79
+ "vocab_size": 32005
80
+ },
81
+ "decoder_start_token_id": 5,
82
+ "early_stopping": true,
83
+ "encoder": {
84
+ "_name_or_path": "camembert-base",
85
+ "add_cross_attention": false,
86
+ "architectures": [
87
+ "CamembertForMaskedLM"
88
+ ],
89
+ "attention_probs_dropout_prob": 0.1,
90
+ "bad_words_ids": null,
91
+ "bos_token_id": 5,
92
+ "chunk_size_feed_forward": 0,
93
+ "decoder_start_token_id": null,
94
+ "diversity_penalty": 0.0,
95
+ "do_sample": false,
96
+ "early_stopping": false,
97
+ "encoder_no_repeat_ngram_size": 0,
98
+ "eos_token_id": 6,
99
+ "finetuning_task": null,
100
+ "forced_bos_token_id": null,
101
+ "forced_eos_token_id": null,
102
+ "gradient_checkpointing": false,
103
+ "hidden_act": "gelu",
104
+ "hidden_dropout_prob": 0.1,
105
+ "hidden_size": 768,
106
+ "id2label": {
107
+ "0": "LABEL_0",
108
+ "1": "LABEL_1"
109
+ },
110
+ "initializer_range": 0.02,
111
+ "intermediate_size": 3072,
112
+ "is_decoder": false,
113
+ "is_encoder_decoder": false,
114
+ "label2id": {
115
+ "LABEL_0": 0,
116
+ "LABEL_1": 1
117
+ },
118
+ "layer_norm_eps": 1e-05,
119
+ "length_penalty": 1.0,
120
+ "max_length": 20,
121
+ "max_position_embeddings": 514,
122
+ "min_length": 0,
123
+ "model_type": "camembert",
124
+ "no_repeat_ngram_size": 0,
125
+ "num_attention_heads": 12,
126
+ "num_beam_groups": 1,
127
+ "num_beams": 1,
128
+ "num_hidden_layers": 12,
129
+ "num_return_sequences": 1,
130
+ "output_attentions": false,
131
+ "output_hidden_states": false,
132
+ "output_past": true,
133
+ "output_scores": false,
134
+ "pad_token_id": 1,
135
+ "position_embedding_type": "absolute",
136
+ "prefix": null,
137
+ "problem_type": null,
138
+ "pruned_heads": {},
139
+ "remove_invalid_values": false,
140
+ "repetition_penalty": 1.0,
141
+ "return_dict": true,
142
+ "return_dict_in_generate": false,
143
+ "sep_token_id": null,
144
+ "task_specific_params": null,
145
+ "temperature": 1.0,
146
+ "tie_encoder_decoder": false,
147
+ "tie_word_embeddings": true,
148
+ "tokenizer_class": null,
149
+ "top_k": 50,
150
+ "top_p": 1.0,
151
+ "torchscript": false,
152
+ "transformers_version": "4.7.0.dev0",
153
+ "type_vocab_size": 1,
154
+ "use_bfloat16": false,
155
+ "use_cache": true,
156
+ "vocab_size": 32005
157
+ },
158
+ "eos_token_id": 6,
159
+ "is_encoder_decoder": true,
160
+ "length_penalty": 2.0,
161
+ "max_length": 64,
162
+ "model_type": "encoder-decoder",
163
+ "no_repeat_ngram_size": 3,
164
+ "num_beams": 4,
165
+ "tie_encoder_decoder": true,
166
+ "transformers_version": null,
167
+ "vocab_size": 32005
168
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:36a2de86ee183c9df5c9fb380193753aa4dd027b96b3d6a5218a1ae5f8664016
3
+ size 1112381587
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e00e2ecc71eec5a8747dff943911b580750fb853339445f74c06097e1ea1bb37
3
+ size 558651685
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:224e0283b9afbef0547fd5c07ee0ae3ae86b7a4a606cff615507ec06c947822b
3
+ size 14657
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4b190a585fe2366c581005a25a7f68c9ada353de6e09e13105e8b1f80466cfb
3
+ size 623
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "special_tokens_map_file": null, "name_or_path": "camembert-base"}
trainer_state.json ADDED
@@ -0,0 +1,1054 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.6920639267619961,
5
+ "global_step": 86500,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 4.997283134350093e-05,
13
+ "loss": 1.6457,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.02,
18
+ "learning_rate": 4.994566268700186e-05,
19
+ "loss": 1.6172,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.03,
24
+ "learning_rate": 4.991849403050279e-05,
25
+ "loss": 1.6618,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.04,
30
+ "learning_rate": 4.989132537400373e-05,
31
+ "loss": 1.6384,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.05,
36
+ "learning_rate": 4.986415671750466e-05,
37
+ "loss": 1.6445,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.06,
42
+ "learning_rate": 4.983698806100559e-05,
43
+ "loss": 1.644,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.07,
48
+ "learning_rate": 4.980981940450652e-05,
49
+ "loss": 1.6763,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.08,
54
+ "learning_rate": 4.9782650748007454e-05,
55
+ "loss": 1.6557,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.09,
60
+ "learning_rate": 4.9755482091508384e-05,
61
+ "loss": 1.6375,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.1,
66
+ "learning_rate": 4.972831343500931e-05,
67
+ "loss": 1.6605,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.11,
72
+ "learning_rate": 4.970114477851024e-05,
73
+ "loss": 1.6571,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 0.12,
78
+ "learning_rate": 4.967397612201118e-05,
79
+ "loss": 1.6556,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 0.13,
84
+ "learning_rate": 4.964680746551211e-05,
85
+ "loss": 1.6593,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 0.14,
90
+ "learning_rate": 4.961963880901304e-05,
91
+ "loss": 1.6578,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 0.15,
96
+ "learning_rate": 4.959247015251397e-05,
97
+ "loss": 1.6519,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 0.16,
102
+ "learning_rate": 4.9565301496014905e-05,
103
+ "loss": 1.6479,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 0.17,
108
+ "learning_rate": 4.9538132839515835e-05,
109
+ "loss": 1.6705,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 0.18,
114
+ "learning_rate": 4.9510964183016765e-05,
115
+ "loss": 1.6807,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 0.19,
120
+ "learning_rate": 4.9483795526517694e-05,
121
+ "loss": 1.6667,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 0.2,
126
+ "learning_rate": 4.945662687001863e-05,
127
+ "loss": 1.6604,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 0.21,
132
+ "learning_rate": 4.942945821351956e-05,
133
+ "loss": 1.6885,
134
+ "step": 10500
135
+ },
136
+ {
137
+ "epoch": 0.22,
138
+ "learning_rate": 4.940228955702049e-05,
139
+ "loss": 1.6648,
140
+ "step": 11000
141
+ },
142
+ {
143
+ "epoch": 0.22,
144
+ "learning_rate": 4.937512090052142e-05,
145
+ "loss": 1.6483,
146
+ "step": 11500
147
+ },
148
+ {
149
+ "epoch": 0.23,
150
+ "learning_rate": 4.9347952244022357e-05,
151
+ "loss": 1.6835,
152
+ "step": 12000
153
+ },
154
+ {
155
+ "epoch": 0.24,
156
+ "learning_rate": 4.9320783587523286e-05,
157
+ "loss": 1.6797,
158
+ "step": 12500
159
+ },
160
+ {
161
+ "epoch": 0.25,
162
+ "learning_rate": 4.9293614931024216e-05,
163
+ "loss": 1.6698,
164
+ "step": 13000
165
+ },
166
+ {
167
+ "epoch": 0.26,
168
+ "learning_rate": 4.9266446274525146e-05,
169
+ "loss": 1.663,
170
+ "step": 13500
171
+ },
172
+ {
173
+ "epoch": 0.27,
174
+ "learning_rate": 4.923927761802608e-05,
175
+ "loss": 1.6552,
176
+ "step": 14000
177
+ },
178
+ {
179
+ "epoch": 0.28,
180
+ "learning_rate": 4.921210896152701e-05,
181
+ "loss": 1.6769,
182
+ "step": 14500
183
+ },
184
+ {
185
+ "epoch": 0.29,
186
+ "learning_rate": 4.918494030502794e-05,
187
+ "loss": 1.6656,
188
+ "step": 15000
189
+ },
190
+ {
191
+ "epoch": 0.3,
192
+ "learning_rate": 4.915777164852887e-05,
193
+ "loss": 1.6515,
194
+ "step": 15500
195
+ },
196
+ {
197
+ "epoch": 0.31,
198
+ "learning_rate": 4.913060299202981e-05,
199
+ "loss": 1.6606,
200
+ "step": 16000
201
+ },
202
+ {
203
+ "epoch": 0.32,
204
+ "learning_rate": 4.910343433553074e-05,
205
+ "loss": 1.6586,
206
+ "step": 16500
207
+ },
208
+ {
209
+ "epoch": 0.33,
210
+ "learning_rate": 4.907626567903167e-05,
211
+ "loss": 1.6945,
212
+ "step": 17000
213
+ },
214
+ {
215
+ "epoch": 0.34,
216
+ "learning_rate": 4.90490970225326e-05,
217
+ "loss": 1.7064,
218
+ "step": 17500
219
+ },
220
+ {
221
+ "epoch": 0.35,
222
+ "learning_rate": 4.9021928366033534e-05,
223
+ "loss": 1.6779,
224
+ "step": 18000
225
+ },
226
+ {
227
+ "epoch": 0.36,
228
+ "learning_rate": 4.899475970953446e-05,
229
+ "loss": 1.6548,
230
+ "step": 18500
231
+ },
232
+ {
233
+ "epoch": 0.37,
234
+ "learning_rate": 4.896759105303539e-05,
235
+ "loss": 1.6983,
236
+ "step": 19000
237
+ },
238
+ {
239
+ "epoch": 0.38,
240
+ "learning_rate": 4.894042239653632e-05,
241
+ "loss": 1.6528,
242
+ "step": 19500
243
+ },
244
+ {
245
+ "epoch": 0.39,
246
+ "learning_rate": 4.891325374003726e-05,
247
+ "loss": 1.6845,
248
+ "step": 20000
249
+ },
250
+ {
251
+ "epoch": 0.4,
252
+ "learning_rate": 4.888608508353818e-05,
253
+ "loss": 1.672,
254
+ "step": 20500
255
+ },
256
+ {
257
+ "epoch": 0.41,
258
+ "learning_rate": 4.885891642703912e-05,
259
+ "loss": 1.6565,
260
+ "step": 21000
261
+ },
262
+ {
263
+ "epoch": 0.42,
264
+ "learning_rate": 4.883174777054005e-05,
265
+ "loss": 1.6549,
266
+ "step": 21500
267
+ },
268
+ {
269
+ "epoch": 0.43,
270
+ "learning_rate": 4.8804579114040985e-05,
271
+ "loss": 1.6702,
272
+ "step": 22000
273
+ },
274
+ {
275
+ "epoch": 0.44,
276
+ "learning_rate": 4.877741045754191e-05,
277
+ "loss": 1.6959,
278
+ "step": 22500
279
+ },
280
+ {
281
+ "epoch": 0.45,
282
+ "learning_rate": 4.8750241801042845e-05,
283
+ "loss": 1.6585,
284
+ "step": 23000
285
+ },
286
+ {
287
+ "epoch": 0.46,
288
+ "learning_rate": 4.8723073144543774e-05,
289
+ "loss": 1.678,
290
+ "step": 23500
291
+ },
292
+ {
293
+ "epoch": 0.47,
294
+ "learning_rate": 4.869590448804471e-05,
295
+ "loss": 1.6927,
296
+ "step": 24000
297
+ },
298
+ {
299
+ "epoch": 0.48,
300
+ "learning_rate": 4.8668735831545634e-05,
301
+ "loss": 1.6876,
302
+ "step": 24500
303
+ },
304
+ {
305
+ "epoch": 0.49,
306
+ "learning_rate": 4.864156717504657e-05,
307
+ "loss": 1.6907,
308
+ "step": 25000
309
+ },
310
+ {
311
+ "epoch": 0.5,
312
+ "learning_rate": 4.86143985185475e-05,
313
+ "loss": 1.6822,
314
+ "step": 25500
315
+ },
316
+ {
317
+ "epoch": 0.51,
318
+ "learning_rate": 4.8587229862048436e-05,
319
+ "loss": 1.7038,
320
+ "step": 26000
321
+ },
322
+ {
323
+ "epoch": 0.52,
324
+ "learning_rate": 4.856006120554936e-05,
325
+ "loss": 1.6671,
326
+ "step": 26500
327
+ },
328
+ {
329
+ "epoch": 0.53,
330
+ "learning_rate": 4.8532892549050296e-05,
331
+ "loss": 1.6546,
332
+ "step": 27000
333
+ },
334
+ {
335
+ "epoch": 0.54,
336
+ "learning_rate": 4.8505723892551226e-05,
337
+ "loss": 1.6896,
338
+ "step": 27500
339
+ },
340
+ {
341
+ "epoch": 0.55,
342
+ "learning_rate": 4.847855523605216e-05,
343
+ "loss": 1.6644,
344
+ "step": 28000
345
+ },
346
+ {
347
+ "epoch": 0.56,
348
+ "learning_rate": 4.8451386579553085e-05,
349
+ "loss": 1.6728,
350
+ "step": 28500
351
+ },
352
+ {
353
+ "epoch": 0.57,
354
+ "learning_rate": 4.842421792305402e-05,
355
+ "loss": 1.687,
356
+ "step": 29000
357
+ },
358
+ {
359
+ "epoch": 0.58,
360
+ "learning_rate": 4.839704926655495e-05,
361
+ "loss": 1.6557,
362
+ "step": 29500
363
+ },
364
+ {
365
+ "epoch": 0.59,
366
+ "learning_rate": 4.836988061005588e-05,
367
+ "loss": 1.7109,
368
+ "step": 30000
369
+ },
370
+ {
371
+ "epoch": 0.6,
372
+ "learning_rate": 4.834271195355681e-05,
373
+ "loss": 1.7037,
374
+ "step": 30500
375
+ },
376
+ {
377
+ "epoch": 0.61,
378
+ "learning_rate": 4.831554329705775e-05,
379
+ "loss": 1.6809,
380
+ "step": 31000
381
+ },
382
+ {
383
+ "epoch": 0.62,
384
+ "learning_rate": 4.828837464055868e-05,
385
+ "loss": 1.6785,
386
+ "step": 31500
387
+ },
388
+ {
389
+ "epoch": 0.63,
390
+ "learning_rate": 4.826120598405961e-05,
391
+ "loss": 1.6842,
392
+ "step": 32000
393
+ },
394
+ {
395
+ "epoch": 0.64,
396
+ "learning_rate": 4.8234037327560536e-05,
397
+ "loss": 1.6569,
398
+ "step": 32500
399
+ },
400
+ {
401
+ "epoch": 0.65,
402
+ "learning_rate": 4.820686867106147e-05,
403
+ "loss": 1.662,
404
+ "step": 33000
405
+ },
406
+ {
407
+ "epoch": 0.66,
408
+ "learning_rate": 4.81797000145624e-05,
409
+ "loss": 1.6857,
410
+ "step": 33500
411
+ },
412
+ {
413
+ "epoch": 0.67,
414
+ "learning_rate": 4.815253135806333e-05,
415
+ "loss": 1.7066,
416
+ "step": 34000
417
+ },
418
+ {
419
+ "epoch": 0.67,
420
+ "learning_rate": 4.812536270156426e-05,
421
+ "loss": 1.6824,
422
+ "step": 34500
423
+ },
424
+ {
425
+ "epoch": 0.68,
426
+ "learning_rate": 4.80981940450652e-05,
427
+ "loss": 1.6847,
428
+ "step": 35000
429
+ },
430
+ {
431
+ "epoch": 0.69,
432
+ "learning_rate": 4.807102538856613e-05,
433
+ "loss": 1.6629,
434
+ "step": 35500
435
+ },
436
+ {
437
+ "epoch": 0.7,
438
+ "learning_rate": 4.804385673206706e-05,
439
+ "loss": 1.6498,
440
+ "step": 36000
441
+ },
442
+ {
443
+ "epoch": 0.71,
444
+ "learning_rate": 4.801668807556799e-05,
445
+ "loss": 1.6833,
446
+ "step": 36500
447
+ },
448
+ {
449
+ "epoch": 0.72,
450
+ "learning_rate": 4.7989519419068924e-05,
451
+ "loss": 1.6693,
452
+ "step": 37000
453
+ },
454
+ {
455
+ "epoch": 0.73,
456
+ "learning_rate": 4.7962350762569854e-05,
457
+ "loss": 1.6912,
458
+ "step": 37500
459
+ },
460
+ {
461
+ "epoch": 0.74,
462
+ "learning_rate": 4.7935182106070784e-05,
463
+ "loss": 1.6591,
464
+ "step": 38000
465
+ },
466
+ {
467
+ "epoch": 0.75,
468
+ "learning_rate": 4.7908013449571714e-05,
469
+ "loss": 1.6655,
470
+ "step": 38500
471
+ },
472
+ {
473
+ "epoch": 0.76,
474
+ "learning_rate": 4.788084479307265e-05,
475
+ "loss": 1.6699,
476
+ "step": 39000
477
+ },
478
+ {
479
+ "epoch": 0.77,
480
+ "learning_rate": 4.785367613657358e-05,
481
+ "loss": 1.6544,
482
+ "step": 39500
483
+ },
484
+ {
485
+ "epoch": 0.78,
486
+ "learning_rate": 4.782650748007451e-05,
487
+ "loss": 1.6876,
488
+ "step": 40000
489
+ },
490
+ {
491
+ "epoch": 0.79,
492
+ "learning_rate": 4.779933882357544e-05,
493
+ "loss": 1.6916,
494
+ "step": 40500
495
+ },
496
+ {
497
+ "epoch": 0.8,
498
+ "learning_rate": 4.7772170167076376e-05,
499
+ "loss": 1.6976,
500
+ "step": 41000
501
+ },
502
+ {
503
+ "epoch": 0.81,
504
+ "learning_rate": 4.7745001510577305e-05,
505
+ "loss": 1.6905,
506
+ "step": 41500
507
+ },
508
+ {
509
+ "epoch": 0.82,
510
+ "learning_rate": 4.7717832854078235e-05,
511
+ "loss": 1.6886,
512
+ "step": 42000
513
+ },
514
+ {
515
+ "epoch": 0.83,
516
+ "learning_rate": 4.7690664197579165e-05,
517
+ "loss": 1.6585,
518
+ "step": 42500
519
+ },
520
+ {
521
+ "epoch": 0.84,
522
+ "learning_rate": 4.76634955410801e-05,
523
+ "loss": 1.6877,
524
+ "step": 43000
525
+ },
526
+ {
527
+ "epoch": 0.85,
528
+ "learning_rate": 4.763632688458103e-05,
529
+ "loss": 1.6786,
530
+ "step": 43500
531
+ },
532
+ {
533
+ "epoch": 0.86,
534
+ "learning_rate": 4.760915822808196e-05,
535
+ "loss": 1.676,
536
+ "step": 44000
537
+ },
538
+ {
539
+ "epoch": 0.87,
540
+ "learning_rate": 4.758198957158289e-05,
541
+ "loss": 1.6702,
542
+ "step": 44500
543
+ },
544
+ {
545
+ "epoch": 0.88,
546
+ "learning_rate": 4.755482091508383e-05,
547
+ "loss": 1.6733,
548
+ "step": 45000
549
+ },
550
+ {
551
+ "epoch": 0.89,
552
+ "learning_rate": 4.752765225858476e-05,
553
+ "loss": 1.6813,
554
+ "step": 45500
555
+ },
556
+ {
557
+ "epoch": 0.9,
558
+ "learning_rate": 4.750048360208569e-05,
559
+ "loss": 1.6726,
560
+ "step": 46000
561
+ },
562
+ {
563
+ "epoch": 0.91,
564
+ "learning_rate": 4.7473314945586616e-05,
565
+ "loss": 1.6844,
566
+ "step": 46500
567
+ },
568
+ {
569
+ "epoch": 0.92,
570
+ "learning_rate": 4.744614628908755e-05,
571
+ "loss": 1.6694,
572
+ "step": 47000
573
+ },
574
+ {
575
+ "epoch": 0.93,
576
+ "learning_rate": 4.741897763258848e-05,
577
+ "loss": 1.6795,
578
+ "step": 47500
579
+ },
580
+ {
581
+ "epoch": 0.94,
582
+ "learning_rate": 4.739180897608941e-05,
583
+ "loss": 1.6512,
584
+ "step": 48000
585
+ },
586
+ {
587
+ "epoch": 0.95,
588
+ "learning_rate": 4.736464031959034e-05,
589
+ "loss": 1.7015,
590
+ "step": 48500
591
+ },
592
+ {
593
+ "epoch": 0.96,
594
+ "learning_rate": 4.733747166309128e-05,
595
+ "loss": 1.6633,
596
+ "step": 49000
597
+ },
598
+ {
599
+ "epoch": 0.97,
600
+ "learning_rate": 4.73103030065922e-05,
601
+ "loss": 1.648,
602
+ "step": 49500
603
+ },
604
+ {
605
+ "epoch": 0.98,
606
+ "learning_rate": 4.728313435009314e-05,
607
+ "loss": 1.6985,
608
+ "step": 50000
609
+ },
610
+ {
611
+ "epoch": 0.99,
612
+ "learning_rate": 4.725596569359407e-05,
613
+ "loss": 1.6694,
614
+ "step": 50500
615
+ },
616
+ {
617
+ "epoch": 1.0,
618
+ "learning_rate": 4.7228797037095004e-05,
619
+ "loss": 1.709,
620
+ "step": 51000
621
+ },
622
+ {
623
+ "epoch": 1.01,
624
+ "learning_rate": 4.720162838059593e-05,
625
+ "loss": 1.5724,
626
+ "step": 51500
627
+ },
628
+ {
629
+ "epoch": 1.02,
630
+ "learning_rate": 4.7174459724096864e-05,
631
+ "loss": 1.5297,
632
+ "step": 52000
633
+ },
634
+ {
635
+ "epoch": 1.03,
636
+ "learning_rate": 4.7147291067597793e-05,
637
+ "loss": 1.5361,
638
+ "step": 52500
639
+ },
640
+ {
641
+ "epoch": 1.04,
642
+ "learning_rate": 4.712012241109873e-05,
643
+ "loss": 1.5141,
644
+ "step": 53000
645
+ },
646
+ {
647
+ "epoch": 1.05,
648
+ "learning_rate": 4.709295375459965e-05,
649
+ "loss": 1.5395,
650
+ "step": 53500
651
+ },
652
+ {
653
+ "epoch": 1.06,
654
+ "learning_rate": 4.706578509810059e-05,
655
+ "loss": 1.5198,
656
+ "step": 54000
657
+ },
658
+ {
659
+ "epoch": 1.07,
660
+ "learning_rate": 4.703861644160152e-05,
661
+ "loss": 1.5742,
662
+ "step": 54500
663
+ },
664
+ {
665
+ "epoch": 1.08,
666
+ "learning_rate": 4.701144778510245e-05,
667
+ "loss": 1.5446,
668
+ "step": 55000
669
+ },
670
+ {
671
+ "epoch": 1.09,
672
+ "learning_rate": 4.698427912860338e-05,
673
+ "loss": 1.5587,
674
+ "step": 55500
675
+ },
676
+ {
677
+ "epoch": 1.1,
678
+ "learning_rate": 4.695711047210431e-05,
679
+ "loss": 1.5487,
680
+ "step": 56000
681
+ },
682
+ {
683
+ "epoch": 1.11,
684
+ "learning_rate": 4.6929941815605245e-05,
685
+ "loss": 1.5434,
686
+ "step": 56500
687
+ },
688
+ {
689
+ "epoch": 1.12,
690
+ "learning_rate": 4.6902773159106175e-05,
691
+ "loss": 1.5424,
692
+ "step": 57000
693
+ },
694
+ {
695
+ "epoch": 1.12,
696
+ "learning_rate": 4.6875604502607104e-05,
697
+ "loss": 1.5343,
698
+ "step": 57500
699
+ },
700
+ {
701
+ "epoch": 1.13,
702
+ "learning_rate": 4.6848435846108034e-05,
703
+ "loss": 1.5479,
704
+ "step": 58000
705
+ },
706
+ {
707
+ "epoch": 1.14,
708
+ "learning_rate": 4.682126718960897e-05,
709
+ "loss": 1.5453,
710
+ "step": 58500
711
+ },
712
+ {
713
+ "epoch": 1.15,
714
+ "learning_rate": 4.67940985331099e-05,
715
+ "loss": 1.5283,
716
+ "step": 59000
717
+ },
718
+ {
719
+ "epoch": 1.16,
720
+ "learning_rate": 4.676692987661083e-05,
721
+ "loss": 1.5689,
722
+ "step": 59500
723
+ },
724
+ {
725
+ "epoch": 1.17,
726
+ "learning_rate": 4.673976122011176e-05,
727
+ "loss": 1.5636,
728
+ "step": 60000
729
+ },
730
+ {
731
+ "epoch": 1.18,
732
+ "learning_rate": 4.6712592563612696e-05,
733
+ "loss": 1.5483,
734
+ "step": 60500
735
+ },
736
+ {
737
+ "epoch": 1.19,
738
+ "learning_rate": 4.6685423907113626e-05,
739
+ "loss": 1.5425,
740
+ "step": 61000
741
+ },
742
+ {
743
+ "epoch": 1.2,
744
+ "learning_rate": 4.6658255250614556e-05,
745
+ "loss": 1.5662,
746
+ "step": 61500
747
+ },
748
+ {
749
+ "epoch": 1.21,
750
+ "learning_rate": 4.6631086594115485e-05,
751
+ "loss": 1.576,
752
+ "step": 62000
753
+ },
754
+ {
755
+ "epoch": 1.22,
756
+ "learning_rate": 4.660391793761642e-05,
757
+ "loss": 1.5613,
758
+ "step": 62500
759
+ },
760
+ {
761
+ "epoch": 1.23,
762
+ "learning_rate": 4.657674928111735e-05,
763
+ "loss": 1.5404,
764
+ "step": 63000
765
+ },
766
+ {
767
+ "epoch": 1.24,
768
+ "learning_rate": 4.654958062461828e-05,
769
+ "loss": 1.5614,
770
+ "step": 63500
771
+ },
772
+ {
773
+ "epoch": 1.25,
774
+ "learning_rate": 4.652241196811921e-05,
775
+ "loss": 1.551,
776
+ "step": 64000
777
+ },
778
+ {
779
+ "epoch": 1.26,
780
+ "learning_rate": 4.649524331162015e-05,
781
+ "loss": 1.5603,
782
+ "step": 64500
783
+ },
784
+ {
785
+ "epoch": 1.27,
786
+ "learning_rate": 4.646807465512108e-05,
787
+ "loss": 1.5794,
788
+ "step": 65000
789
+ },
790
+ {
791
+ "epoch": 1.28,
792
+ "learning_rate": 4.644090599862201e-05,
793
+ "loss": 1.5551,
794
+ "step": 65500
795
+ },
796
+ {
797
+ "epoch": 1.29,
798
+ "learning_rate": 4.641373734212294e-05,
799
+ "loss": 1.5965,
800
+ "step": 66000
801
+ },
802
+ {
803
+ "epoch": 1.3,
804
+ "learning_rate": 4.638656868562387e-05,
805
+ "loss": 1.5662,
806
+ "step": 66500
807
+ },
808
+ {
809
+ "epoch": 1.31,
810
+ "learning_rate": 4.63594000291248e-05,
811
+ "loss": 1.6079,
812
+ "step": 67000
813
+ },
814
+ {
815
+ "epoch": 1.32,
816
+ "learning_rate": 4.633223137262573e-05,
817
+ "loss": 1.5807,
818
+ "step": 67500
819
+ },
820
+ {
821
+ "epoch": 1.33,
822
+ "learning_rate": 4.630506271612666e-05,
823
+ "loss": 1.5592,
824
+ "step": 68000
825
+ },
826
+ {
827
+ "epoch": 1.34,
828
+ "learning_rate": 4.62778940596276e-05,
829
+ "loss": 1.5741,
830
+ "step": 68500
831
+ },
832
+ {
833
+ "epoch": 1.35,
834
+ "learning_rate": 4.625072540312852e-05,
835
+ "loss": 1.5827,
836
+ "step": 69000
837
+ },
838
+ {
839
+ "epoch": 1.36,
840
+ "learning_rate": 4.622355674662946e-05,
841
+ "loss": 1.5656,
842
+ "step": 69500
843
+ },
844
+ {
845
+ "epoch": 1.37,
846
+ "learning_rate": 4.619638809013039e-05,
847
+ "loss": 1.5764,
848
+ "step": 70000
849
+ },
850
+ {
851
+ "epoch": 1.38,
852
+ "learning_rate": 4.6169219433631325e-05,
853
+ "loss": 1.556,
854
+ "step": 70500
855
+ },
856
+ {
857
+ "epoch": 1.39,
858
+ "learning_rate": 4.614205077713225e-05,
859
+ "loss": 1.582,
860
+ "step": 71000
861
+ },
862
+ {
863
+ "epoch": 1.4,
864
+ "learning_rate": 4.6114882120633184e-05,
865
+ "loss": 1.5789,
866
+ "step": 71500
867
+ },
868
+ {
869
+ "epoch": 1.41,
870
+ "learning_rate": 4.6087713464134114e-05,
871
+ "loss": 1.5719,
872
+ "step": 72000
873
+ },
874
+ {
875
+ "epoch": 1.42,
876
+ "learning_rate": 4.606054480763505e-05,
877
+ "loss": 1.581,
878
+ "step": 72500
879
+ },
880
+ {
881
+ "epoch": 1.43,
882
+ "learning_rate": 4.603337615113597e-05,
883
+ "loss": 1.5885,
884
+ "step": 73000
885
+ },
886
+ {
887
+ "epoch": 1.44,
888
+ "learning_rate": 4.600620749463691e-05,
889
+ "loss": 1.5516,
890
+ "step": 73500
891
+ },
892
+ {
893
+ "epoch": 1.45,
894
+ "learning_rate": 4.597903883813784e-05,
895
+ "loss": 1.59,
896
+ "step": 74000
897
+ },
898
+ {
899
+ "epoch": 1.46,
900
+ "learning_rate": 4.5951870181638776e-05,
901
+ "loss": 1.5968,
902
+ "step": 74500
903
+ },
904
+ {
905
+ "epoch": 1.47,
906
+ "learning_rate": 4.59247015251397e-05,
907
+ "loss": 1.5865,
908
+ "step": 75000
909
+ },
910
+ {
911
+ "epoch": 1.48,
912
+ "learning_rate": 4.5897532868640636e-05,
913
+ "loss": 1.5682,
914
+ "step": 75500
915
+ },
916
+ {
917
+ "epoch": 1.49,
918
+ "learning_rate": 4.5870364212141565e-05,
919
+ "loss": 1.5711,
920
+ "step": 76000
921
+ },
922
+ {
923
+ "epoch": 1.5,
924
+ "learning_rate": 4.58431955556425e-05,
925
+ "loss": 1.5595,
926
+ "step": 76500
927
+ },
928
+ {
929
+ "epoch": 1.51,
930
+ "learning_rate": 4.5816026899143425e-05,
931
+ "loss": 1.5898,
932
+ "step": 77000
933
+ },
934
+ {
935
+ "epoch": 1.52,
936
+ "learning_rate": 4.578885824264436e-05,
937
+ "loss": 1.6049,
938
+ "step": 77500
939
+ },
940
+ {
941
+ "epoch": 1.53,
942
+ "learning_rate": 4.576168958614529e-05,
943
+ "loss": 1.613,
944
+ "step": 78000
945
+ },
946
+ {
947
+ "epoch": 1.54,
948
+ "learning_rate": 4.573452092964622e-05,
949
+ "loss": 1.5761,
950
+ "step": 78500
951
+ },
952
+ {
953
+ "epoch": 1.55,
954
+ "learning_rate": 4.570735227314715e-05,
955
+ "loss": 1.5901,
956
+ "step": 79000
957
+ },
958
+ {
959
+ "epoch": 1.56,
960
+ "learning_rate": 4.568018361664809e-05,
961
+ "loss": 1.6102,
962
+ "step": 79500
963
+ },
964
+ {
965
+ "epoch": 1.56,
966
+ "learning_rate": 4.565301496014902e-05,
967
+ "loss": 1.5958,
968
+ "step": 80000
969
+ },
970
+ {
971
+ "epoch": 1.57,
972
+ "learning_rate": 4.5625846303649946e-05,
973
+ "loss": 1.5903,
974
+ "step": 80500
975
+ },
976
+ {
977
+ "epoch": 1.58,
978
+ "learning_rate": 4.5598677647150876e-05,
979
+ "loss": 1.5798,
980
+ "step": 81000
981
+ },
982
+ {
983
+ "epoch": 1.59,
984
+ "learning_rate": 4.557150899065181e-05,
985
+ "loss": 1.5832,
986
+ "step": 81500
987
+ },
988
+ {
989
+ "epoch": 1.6,
990
+ "learning_rate": 4.554434033415274e-05,
991
+ "loss": 1.6358,
992
+ "step": 82000
993
+ },
994
+ {
995
+ "epoch": 1.61,
996
+ "learning_rate": 4.551717167765367e-05,
997
+ "loss": 1.5969,
998
+ "step": 82500
999
+ },
1000
+ {
1001
+ "epoch": 1.62,
1002
+ "learning_rate": 4.54900030211546e-05,
1003
+ "loss": 1.6018,
1004
+ "step": 83000
1005
+ },
1006
+ {
1007
+ "epoch": 1.63,
1008
+ "learning_rate": 4.546283436465554e-05,
1009
+ "loss": 1.5919,
1010
+ "step": 83500
1011
+ },
1012
+ {
1013
+ "epoch": 1.64,
1014
+ "learning_rate": 4.543566570815647e-05,
1015
+ "loss": 1.5925,
1016
+ "step": 84000
1017
+ },
1018
+ {
1019
+ "epoch": 1.65,
1020
+ "learning_rate": 4.54084970516574e-05,
1021
+ "loss": 1.6047,
1022
+ "step": 84500
1023
+ },
1024
+ {
1025
+ "epoch": 1.66,
1026
+ "learning_rate": 4.538132839515833e-05,
1027
+ "loss": 1.6178,
1028
+ "step": 85000
1029
+ },
1030
+ {
1031
+ "epoch": 1.67,
1032
+ "learning_rate": 4.5354159738659264e-05,
1033
+ "loss": 1.5892,
1034
+ "step": 85500
1035
+ },
1036
+ {
1037
+ "epoch": 1.68,
1038
+ "learning_rate": 4.5326991082160194e-05,
1039
+ "loss": 1.6217,
1040
+ "step": 86000
1041
+ },
1042
+ {
1043
+ "epoch": 1.69,
1044
+ "learning_rate": 4.5299822425661123e-05,
1045
+ "loss": 1.5807,
1046
+ "step": 86500
1047
+ }
1048
+ ],
1049
+ "max_steps": 920178,
1050
+ "num_train_epochs": 18,
1051
+ "total_flos": 1930009185792000.0,
1052
+ "trial_name": null,
1053
+ "trial_params": null
1054
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:240d71a042455be772148965f0bf68f00aabba3cded4bd95bc744072787219d2
3
+ size 2479
unigram.json ADDED
The diff for this file is too large to render. See raw diff