Hugging Face's logo

Glavin001
/
coqar-questions-llama-2-7b-v0.1

Text Generation
Transformers
PyTorch
English
llama
text-generation-inference
Inference Endpoints
Model card Files Community
Glavin001 commited on
Commit
49f105d
·
1 Parent(s): ebb8495

Upload folder using huggingface_hub

Browse files
Files changed (11) hide show
  1. checkpoints/checkpoint-258/README.md +21 -0
  2. checkpoints/checkpoint-258/adapter_config.json +26 -0
  3. checkpoints/checkpoint-258/adapter_model.bin +3 -0
  4. checkpoints/checkpoint-258/adapter_model/README.md +21 -0
  5. checkpoints/checkpoint-258/adapter_model/adapter_config.json +26 -0
  6. checkpoints/checkpoint-258/adapter_model/adapter_model.bin +3 -0
  7. checkpoints/checkpoint-258/optimizer.pt +3 -0
  8. checkpoints/checkpoint-258/rng_state.pth +3 -0
  9. checkpoints/checkpoint-258/scheduler.pt +3 -0
  10. checkpoints/checkpoint-258/trainer_state.json +1663 -0
  11. checkpoints/checkpoint-258/training_args.bin +3 -0
checkpoints/checkpoint-258/README.md ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - quant_method: bitsandbytes
9
+ - load_in_8bit: True
10
+ - load_in_4bit: False
11
+ - llm_int8_threshold: 6.0
12
+ - llm_int8_skip_modules: None
13
+ - llm_int8_enable_fp32_cpu_offload: False
14
+ - llm_int8_has_fp16_weight: False
15
+ - bnb_4bit_quant_type: fp4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float32
18
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.6.0.dev0
checkpoints/checkpoint-258/adapter_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
4
+ "bias": "none",
5
+ "fan_in_fan_out": null,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 16,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 32,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "o_proj",
18
+ "k_proj",
19
+ "v_proj",
20
+ "q_proj",
21
+ "gate_proj",
22
+ "down_proj",
23
+ "up_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM"
26
+ }
checkpoints/checkpoint-258/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33694da367495220ba704ce85f33c2b863ac8926247f019890b4f9b8cce4df61
3
+ size 319977229
checkpoints/checkpoint-258/adapter_model/README.md ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - quant_method: bitsandbytes
9
+ - load_in_8bit: True
10
+ - load_in_4bit: False
11
+ - llm_int8_threshold: 6.0
12
+ - llm_int8_skip_modules: None
13
+ - llm_int8_enable_fp32_cpu_offload: False
14
+ - llm_int8_has_fp16_weight: False
15
+ - bnb_4bit_quant_type: fp4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float32
18
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.6.0.dev0
checkpoints/checkpoint-258/adapter_model/adapter_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
4
+ "bias": "none",
5
+ "fan_in_fan_out": null,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 16,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 32,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "o_proj",
18
+ "k_proj",
19
+ "v_proj",
20
+ "q_proj",
21
+ "gate_proj",
22
+ "down_proj",
23
+ "up_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM"
26
+ }
checkpoints/checkpoint-258/adapter_model/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33694da367495220ba704ce85f33c2b863ac8926247f019890b4f9b8cce4df61
3
+ size 319977229
checkpoints/checkpoint-258/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cffded315cbf135fdba60e950d39ca68d921480ef4b0e575b44b038693407aa0
3
+ size 160736095
checkpoints/checkpoint-258/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2368bf51309cc1094f7890994b76045a2d12b581dfdca5f7f5cedbf1fd4a3f47
3
+ size 14575
checkpoints/checkpoint-258/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:436404ee3cdd61b40274ab45342672e13c20a7c00708b66c47dfceb5c10e2a19
3
+ size 627
checkpoints/checkpoint-258/trainer_state.json ADDED
@@ -0,0 +1,1663 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 20,
6
+ "global_step": 258,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 2e-05,
14
+ "loss": 0.9238,
15
+ "step": 1
16
+ },
17
+ {
18
+ "epoch": 0.02,
19
+ "learning_rate": 4e-05,
20
+ "loss": 0.886,
21
+ "step": 2
22
+ },
23
+ {
24
+ "epoch": 0.03,
25
+ "learning_rate": 6e-05,
26
+ "loss": 0.883,
27
+ "step": 3
28
+ },
29
+ {
30
+ "epoch": 0.05,
31
+ "learning_rate": 8e-05,
32
+ "loss": 0.8953,
33
+ "step": 4
34
+ },
35
+ {
36
+ "epoch": 0.06,
37
+ "learning_rate": 0.0001,
38
+ "loss": 0.9026,
39
+ "step": 5
40
+ },
41
+ {
42
+ "epoch": 0.07,
43
+ "learning_rate": 0.00012,
44
+ "loss": 0.836,
45
+ "step": 6
46
+ },
47
+ {
48
+ "epoch": 0.08,
49
+ "learning_rate": 0.00014,
50
+ "loss": 0.8306,
51
+ "step": 7
52
+ },
53
+ {
54
+ "epoch": 0.09,
55
+ "learning_rate": 0.00016,
56
+ "loss": 0.8028,
57
+ "step": 8
58
+ },
59
+ {
60
+ "epoch": 0.1,
61
+ "learning_rate": 0.00018,
62
+ "loss": 0.8415,
63
+ "step": 9
64
+ },
65
+ {
66
+ "epoch": 0.12,
67
+ "learning_rate": 0.0002,
68
+ "loss": 0.7948,
69
+ "step": 10
70
+ },
71
+ {
72
+ "epoch": 0.13,
73
+ "learning_rate": 0.00019999083173529673,
74
+ "loss": 0.7679,
75
+ "step": 11
76
+ },
77
+ {
78
+ "epoch": 0.14,
79
+ "learning_rate": 0.0001999633286223284,
80
+ "loss": 0.7847,
81
+ "step": 12
82
+ },
83
+ {
84
+ "epoch": 0.15,
85
+ "learning_rate": 0.00019991749570421146,
86
+ "loss": 0.7691,
87
+ "step": 13
88
+ },
89
+ {
90
+ "epoch": 0.16,
91
+ "learning_rate": 0.00019985334138511237,
92
+ "loss": 0.7327,
93
+ "step": 14
94
+ },
95
+ {
96
+ "epoch": 0.17,
97
+ "learning_rate": 0.0001997708774287068,
98
+ "loss": 0.7505,
99
+ "step": 15
100
+ },
101
+ {
102
+ "epoch": 0.19,
103
+ "learning_rate": 0.0001996701189560223,
104
+ "loss": 0.7268,
105
+ "step": 16
106
+ },
107
+ {
108
+ "epoch": 0.2,
109
+ "learning_rate": 0.00019955108444266585,
110
+ "loss": 0.7068,
111
+ "step": 17
112
+ },
113
+ {
114
+ "epoch": 0.21,
115
+ "learning_rate": 0.00019941379571543596,
116
+ "loss": 0.7,
117
+ "step": 18
118
+ },
119
+ {
120
+ "epoch": 0.22,
121
+ "learning_rate": 0.00019925827794832056,
122
+ "loss": 0.7068,
123
+ "step": 19
124
+ },
125
+ {
126
+ "epoch": 0.23,
127
+ "learning_rate": 0.00019908455965788067,
128
+ "loss": 0.7597,
129
+ "step": 20
130
+ },
131
+ {
132
+ "epoch": 0.23,
133
+ "eval_loss": 0.6609811782836914,
134
+ "eval_runtime": 2.1763,
135
+ "eval_samples_per_second": 13.785,
136
+ "eval_steps_per_second": 6.893,
137
+ "step": 20
138
+ },
139
+ {
140
+ "epoch": 0.24,
141
+ "learning_rate": 0.00019889267269802176,
142
+ "loss": 0.7521,
143
+ "step": 21
144
+ },
145
+ {
146
+ "epoch": 0.26,
147
+ "learning_rate": 0.00019868265225415265,
148
+ "loss": 0.7083,
149
+ "step": 22
150
+ },
151
+ {
152
+ "epoch": 0.27,
153
+ "learning_rate": 0.00019845453683673368,
154
+ "loss": 0.6528,
155
+ "step": 23
156
+ },
157
+ {
158
+ "epoch": 0.28,
159
+ "learning_rate": 0.0001982083682742156,
160
+ "loss": 0.7004,
161
+ "step": 24
162
+ },
163
+ {
164
+ "epoch": 0.29,
165
+ "learning_rate": 0.00019794419170536916,
166
+ "loss": 0.667,
167
+ "step": 25
168
+ },
169
+ {
170
+ "epoch": 0.3,
171
+ "learning_rate": 0.00019766205557100868,
172
+ "loss": 0.7224,
173
+ "step": 26
174
+ },
175
+ {
176
+ "epoch": 0.31,
177
+ "learning_rate": 0.00019736201160510931,
178
+ "loss": 0.6855,
179
+ "step": 27
180
+ },
181
+ {
182
+ "epoch": 0.33,
183
+ "learning_rate": 0.00019704411482532116,
184
+ "loss": 0.6642,
185
+ "step": 28
186
+ },
187
+ {
188
+ "epoch": 0.34,
189
+ "learning_rate": 0.0001967084235228807,
190
+ "loss": 0.6755,
191
+ "step": 29
192
+ },
193
+ {
194
+ "epoch": 0.35,
195
+ "learning_rate": 0.0001963549992519223,
196
+ "loss": 0.6875,
197
+ "step": 30
198
+ },
199
+ {
200
+ "epoch": 0.36,
201
+ "learning_rate": 0.0001959839068181914,
202
+ "loss": 0.6932,
203
+ "step": 31
204
+ },
205
+ {
206
+ "epoch": 0.37,
207
+ "learning_rate": 0.00019559521426716118,
208
+ "loss": 0.6879,
209
+ "step": 32
210
+ },
211
+ {
212
+ "epoch": 0.38,
213
+ "learning_rate": 0.00019518899287155556,
214
+ "loss": 0.6903,
215
+ "step": 33
216
+ },
217
+ {
218
+ "epoch": 0.4,
219
+ "learning_rate": 0.00019476531711828027,
220
+ "loss": 0.6807,
221
+ "step": 34
222
+ },
223
+ {
224
+ "epoch": 0.41,
225
+ "learning_rate": 0.0001943242646947643,
226
+ "loss": 0.6842,
227
+ "step": 35
228
+ },
229
+ {
230
+ "epoch": 0.42,
231
+ "learning_rate": 0.00019386591647471506,
232
+ "loss": 0.7095,
233
+ "step": 36
234
+ },
235
+ {
236
+ "epoch": 0.43,
237
+ "learning_rate": 0.00019339035650328869,
238
+ "loss": 0.6779,
239
+ "step": 37
240
+ },
241
+ {
242
+ "epoch": 0.44,
243
+ "learning_rate": 0.00019289767198167916,
244
+ "loss": 0.6922,
245
+ "step": 38
246
+ },
247
+ {
248
+ "epoch": 0.45,
249
+ "learning_rate": 0.0001923879532511287,
250
+ "loss": 0.684,
251
+ "step": 39
252
+ },
253
+ {
254
+ "epoch": 0.47,
255
+ "learning_rate": 0.0001918612937763622,
256
+ "loss": 0.6737,
257
+ "step": 40
258
+ },
259
+ {
260
+ "epoch": 0.47,
261
+ "eval_loss": 0.6172597408294678,
262
+ "eval_runtime": 2.21,
263
+ "eval_samples_per_second": 13.575,
264
+ "eval_steps_per_second": 6.787,
265
+ "step": 40
266
+ },
267
+ {
268
+ "epoch": 0.48,
269
+ "learning_rate": 0.00019131779012844912,
270
+ "loss": 0.692,
271
+ "step": 41
272
+ },
273
+ {
274
+ "epoch": 0.49,
275
+ "learning_rate": 0.00019075754196709572,
276
+ "loss": 0.639,
277
+ "step": 42
278
+ },
279
+ {
280
+ "epoch": 0.5,
281
+ "learning_rate": 0.00019018065202237083,
282
+ "loss": 0.6448,
283
+ "step": 43
284
+ },
285
+ {
286
+ "epoch": 0.51,
287
+ "learning_rate": 0.0001895872260758688,
288
+ "loss": 0.6595,
289
+ "step": 44
290
+ },
291
+ {
292
+ "epoch": 0.52,
293
+ "learning_rate": 0.00018897737294131284,
294
+ "loss": 0.7268,
295
+ "step": 45
296
+ },
297
+ {
298
+ "epoch": 0.53,
299
+ "learning_rate": 0.0001883512044446023,
300
+ "loss": 0.676,
301
+ "step": 46
302
+ },
303
+ {
304
+ "epoch": 0.55,
305
+ "learning_rate": 0.0001877088354033077,
306
+ "loss": 0.687,
307
+ "step": 47
308
+ },
309
+ {
310
+ "epoch": 0.56,
311
+ "learning_rate": 0.0001870503836056172,
312
+ "loss": 0.649,
313
+ "step": 48
314
+ },
315
+ {
316
+ "epoch": 0.57,
317
+ "learning_rate": 0.00018637596978873835,
318
+ "loss": 0.6888,
319
+ "step": 49
320
+ },
321
+ {
322
+ "epoch": 0.58,
323
+ "learning_rate": 0.00018568571761675893,
324
+ "loss": 0.6602,
325
+ "step": 50
326
+ },
327
+ {
328
+ "epoch": 0.59,
329
+ "learning_rate": 0.0001849797536579715,
330
+ "loss": 0.7303,
331
+ "step": 51
332
+ },
333
+ {
334
+ "epoch": 0.6,
335
+ "learning_rate": 0.0001842582073616649,
336
+ "loss": 0.6195,
337
+ "step": 52
338
+ },
339
+ {
340
+ "epoch": 0.62,
341
+ "learning_rate": 0.000183521211034388,
342
+ "loss": 0.6736,
343
+ "step": 53
344
+ },
345
+ {
346
+ "epoch": 0.63,
347
+ "learning_rate": 0.00018276889981568906,
348
+ "loss": 0.6672,
349
+ "step": 54
350
+ },
351
+ {
352
+ "epoch": 0.64,
353
+ "learning_rate": 0.0001820014116533359,
354
+ "loss": 0.6474,
355
+ "step": 55
356
+ },
357
+ {
358
+ "epoch": 0.65,
359
+ "learning_rate": 0.00018121888727802113,
360
+ "loss": 0.6576,
361
+ "step": 56
362
+ },
363
+ {
364
+ "epoch": 0.66,
365
+ "learning_rate": 0.0001804214701775569,
366
+ "loss": 0.7122,
367
+ "step": 57
368
+ },
369
+ {
370
+ "epoch": 0.67,
371
+ "learning_rate": 0.00017960930657056438,
372
+ "loss": 0.6327,
373
+ "step": 58
374
+ },
375
+ {
376
+ "epoch": 0.69,
377
+ "learning_rate": 0.00017878254537966216,
378
+ "loss": 0.7047,
379
+ "step": 59
380
+ },
381
+ {
382
+ "epoch": 0.7,
383
+ "learning_rate": 0.00017794133820415916,
384
+ "loss": 0.7,
385
+ "step": 60
386
+ },
387
+ {
388
+ "epoch": 0.7,
389
+ "eval_loss": 0.5973738431930542,
390
+ "eval_runtime": 2.1866,
391
+ "eval_samples_per_second": 13.72,
392
+ "eval_steps_per_second": 6.86,
393
+ "step": 60
394
+ },
395
+ {
396
+ "epoch": 0.71,
397
+ "learning_rate": 0.0001770858392922565,
398
+ "loss": 0.6418,
399
+ "step": 61
400
+ },
401
+ {
402
+ "epoch": 0.72,
403
+ "learning_rate": 0.00017621620551276366,
404
+ "loss": 0.6977,
405
+ "step": 62
406
+ },
407
+ {
408
+ "epoch": 0.73,
409
+ "learning_rate": 0.00017533259632633442,
410
+ "loss": 0.6916,
411
+ "step": 63
412
+ },
413
+ {
414
+ "epoch": 0.74,
415
+ "learning_rate": 0.00017443517375622704,
416
+ "loss": 0.6809,
417
+ "step": 64
418
+ },
419
+ {
420
+ "epoch": 0.76,
421
+ "learning_rate": 0.00017352410235859503,
422
+ "loss": 0.6867,
423
+ "step": 65
424
+ },
425
+ {
426
+ "epoch": 0.77,
427
+ "learning_rate": 0.0001725995491923131,
428
+ "loss": 0.6446,
429
+ "step": 66
430
+ },
431
+ {
432
+ "epoch": 0.78,
433
+ "learning_rate": 0.00017166168378834448,
434
+ "loss": 0.6844,
435
+ "step": 67
436
+ },
437
+ {
438
+ "epoch": 0.79,
439
+ "learning_rate": 0.00017071067811865476,
440
+ "loss": 0.6529,
441
+ "step": 68
442
+ },
443
+ {
444
+ "epoch": 0.8,
445
+ "learning_rate": 0.00016974670656467824,
446
+ "loss": 0.7094,
447
+ "step": 69
448
+ },
449
+ {
450
+ "epoch": 0.81,
451
+ "learning_rate": 0.00016876994588534234,
452
+ "loss": 0.7008,
453
+ "step": 70
454
+ },
455
+ {
456
+ "epoch": 0.83,
457
+ "learning_rate": 0.0001677805751846563,
458
+ "loss": 0.6635,
459
+ "step": 71
460
+ },
461
+ {
462
+ "epoch": 0.84,
463
+ "learning_rate": 0.00016677877587886956,
464
+ "loss": 0.6514,
465
+ "step": 72
466
+ },
467
+ {
468
+ "epoch": 0.85,
469
+ "learning_rate": 0.00016576473166320644,
470
+ "loss": 0.6503,
471
+ "step": 73
472
+ },
473
+ {
474
+ "epoch": 0.86,
475
+ "learning_rate": 0.00016473862847818277,
476
+ "loss": 0.7369,
477
+ "step": 74
478
+ },
479
+ {
480
+ "epoch": 0.87,
481
+ "learning_rate": 0.00016370065447551078,
482
+ "loss": 0.6765,
483
+ "step": 75
484
+ },
485
+ {
486
+ "epoch": 0.88,
487
+ "learning_rate": 0.00016265099998359866,
488
+ "loss": 0.6762,
489
+ "step": 76
490
+ },
491
+ {
492
+ "epoch": 0.9,
493
+ "learning_rate": 0.00016158985747265108,
494
+ "loss": 0.669,
495
+ "step": 77
496
+ },
497
+ {
498
+ "epoch": 0.91,
499
+ "learning_rate": 0.00016051742151937655,
500
+ "loss": 0.6726,
501
+ "step": 78
502
+ },
503
+ {
504
+ "epoch": 0.92,
505
+ "learning_rate": 0.000159433888771309,
506
+ "loss": 0.6289,
507
+ "step": 79
508
+ },
509
+ {
510
+ "epoch": 0.93,
511
+ "learning_rate": 0.00015833945791074943,
512
+ "loss": 0.6681,
513
+ "step": 80
514
+ },
515
+ {
516
+ "epoch": 0.93,
517
+ "eval_loss": 0.5905855298042297,
518
+ "eval_runtime": 2.188,
519
+ "eval_samples_per_second": 13.711,
520
+ "eval_steps_per_second": 6.856,
521
+ "step": 80
522
+ },
523
+ {
524
+ "epoch": 0.94,
525
+ "learning_rate": 0.0001572343296183344,
526
+ "loss": 0.6897,
527
+ "step": 81
528
+ },
529
+ {
530
+ "epoch": 0.95,
531
+ "learning_rate": 0.00015611870653623825,
532
+ "loss": 0.6846,
533
+ "step": 82
534
+ },
535
+ {
536
+ "epoch": 0.97,
537
+ "learning_rate": 0.0001549927932310155,
538
+ "loss": 0.7051,
539
+ "step": 83
540
+ },
541
+ {
542
+ "epoch": 0.98,
543
+ "learning_rate": 0.00015385679615609042,
544
+ "loss": 0.6657,
545
+ "step": 84
546
+ },
547
+ {
548
+ "epoch": 0.99,
549
+ "learning_rate": 0.00015271092361390077,
550
+ "loss": 0.6343,
551
+ "step": 85
552
+ },
553
+ {
554
+ "epoch": 1.0,
555
+ "learning_rate": 0.00015155538571770218,
556
+ "loss": 0.6161,
557
+ "step": 86
558
+ },
559
+ {
560
+ "epoch": 1.01,
561
+ "learning_rate": 0.00015039039435304078,
562
+ "loss": 0.6458,
563
+ "step": 87
564
+ },
565
+ {
566
+ "epoch": 1.02,
567
+ "learning_rate": 0.00014921616313890072,
568
+ "loss": 0.6567,
569
+ "step": 88
570
+ },
571
+ {
572
+ "epoch": 1.03,
573
+ "learning_rate": 0.00014803290738853395,
574
+ "loss": 0.6508,
575
+ "step": 89
576
+ },
577
+ {
578
+ "epoch": 1.05,
579
+ "learning_rate": 0.00014684084406997903,
580
+ "loss": 0.6397,
581
+ "step": 90
582
+ },
583
+ {
584
+ "epoch": 1.06,
585
+ "learning_rate": 0.0001456401917662769,
586
+ "loss": 0.6167,
587
+ "step": 91
588
+ },
589
+ {
590
+ "epoch": 1.07,
591
+ "learning_rate": 0.00014443117063539038,
592
+ "loss": 0.6272,
593
+ "step": 92
594
+ },
595
+ {
596
+ "epoch": 1.08,
597
+ "learning_rate": 0.00014321400236983457,
598
+ "loss": 0.6162,
599
+ "step": 93
600
+ },
601
+ {
602
+ "epoch": 1.09,
603
+ "learning_rate": 0.00014198891015602646,
604
+ "loss": 0.6361,
605
+ "step": 94
606
+ },
607
+ {
608
+ "epoch": 1.1,
609
+ "learning_rate": 0.0001407561186333601,
610
+ "loss": 0.635,
611
+ "step": 95
612
+ },
613
+ {
614
+ "epoch": 1.12,
615
+ "learning_rate": 0.00013951585385301555,
616
+ "loss": 0.6452,
617
+ "step": 96
618
+ },
619
+ {
620
+ "epoch": 1.13,
621
+ "learning_rate": 0.000138268343236509,
622
+ "loss": 0.645,
623
+ "step": 97
624
+ },
625
+ {
626
+ "epoch": 1.14,
627
+ "learning_rate": 0.00013701381553399145,
628
+ "loss": 0.6266,
629
+ "step": 98
630
+ },
631
+ {
632
+ "epoch": 1.15,
633
+ "learning_rate": 0.000135752500782304,
634
+ "loss": 0.6294,
635
+ "step": 99
636
+ },
637
+ {
638
+ "epoch": 1.16,
639
+ "learning_rate": 0.00013448463026279704,
640
+ "loss": 0.6133,
641
+ "step": 100
642
+ },
643
+ {
644
+ "epoch": 1.16,
645
+ "eval_loss": 0.5831260681152344,
646
+ "eval_runtime": 2.1769,
647
+ "eval_samples_per_second": 13.781,
648
+ "eval_steps_per_second": 6.891,
649
+ "step": 100
650
+ },
651
+ {
652
+ "epoch": 1.17,
653
+ "learning_rate": 0.0001332104364589212,
654
+ "loss": 0.6629,
655
+ "step": 101
656
+ },
657
+ {
658
+ "epoch": 1.19,
659
+ "learning_rate": 0.000131930153013598,
660
+ "loss": 0.6885,
661
+ "step": 102
662
+ },
663
+ {
664
+ "epoch": 1.2,
665
+ "learning_rate": 0.00013064401468637792,
666
+ "loss": 0.636,
667
+ "step": 103
668
+ },
669
+ {
670
+ "epoch": 1.21,
671
+ "learning_rate": 0.00012935225731039348,
672
+ "loss": 0.6584,
673
+ "step": 104
674
+ },
675
+ {
676
+ "epoch": 1.22,
677
+ "learning_rate": 0.00012805511774911584,
678
+ "loss": 0.6176,
679
+ "step": 105
680
+ },
681
+ {
682
+ "epoch": 1.23,
683
+ "learning_rate": 0.00012675283385292212,
684
+ "loss": 0.6084,
685
+ "step": 106
686
+ },
687
+ {
688
+ "epoch": 1.24,
689
+ "learning_rate": 0.00012544564441548182,
690
+ "loss": 0.6359,
691
+ "step": 107
692
+ },
693
+ {
694
+ "epoch": 1.26,
695
+ "learning_rate": 0.00012413378912997058,
696
+ "loss": 0.6501,
697
+ "step": 108
698
+ },
699
+ {
700
+ "epoch": 1.27,
701
+ "learning_rate": 0.0001228175085451186,
702
+ "loss": 0.6585,
703
+ "step": 109
704
+ },
705
+ {
706
+ "epoch": 1.28,
707
+ "learning_rate": 0.00012149704402110243,
708
+ "loss": 0.6418,
709
+ "step": 110
710
+ },
711
+ {
712
+ "epoch": 1.29,
713
+ "learning_rate": 0.00012017263768528775,
714
+ "loss": 0.6511,
715
+ "step": 111
716
+ },
717
+ {
718
+ "epoch": 1.3,
719
+ "learning_rate": 0.00011884453238783185,
720
+ "loss": 0.617,
721
+ "step": 112
722
+ },
723
+ {
724
+ "epoch": 1.31,
725
+ "learning_rate": 0.00011751297165715309,
726
+ "loss": 0.6165,
727
+ "step": 113
728
+ },
729
+ {
730
+ "epoch": 1.33,
731
+ "learning_rate": 0.0001161781996552765,
732
+ "loss": 0.6143,
733
+ "step": 114
734
+ },
735
+ {
736
+ "epoch": 1.34,
737
+ "learning_rate": 0.00011484046113306262,
738
+ "loss": 0.6743,
739
+ "step": 115
740
+ },
741
+ {
742
+ "epoch": 1.35,
743
+ "learning_rate": 0.00011350000138532902,
744
+ "loss": 0.627,
745
+ "step": 116
746
+ },
747
+ {
748
+ "epoch": 1.36,
749
+ "learning_rate": 0.00011215706620587149,
750
+ "loss": 0.6268,
751
+ "step": 117
752
+ },
753
+ {
754
+ "epoch": 1.37,
755
+ "learning_rate": 0.00011081190184239419,
756
+ "loss": 0.6279,
757
+ "step": 118
758
+ },
759
+ {
760
+ "epoch": 1.38,
761
+ "learning_rate": 0.0001094647549513561,
762
+ "loss": 0.6622,
763
+ "step": 119
764
+ },
765
+ {
766
+ "epoch": 1.4,
767
+ "learning_rate": 0.00010811587255274313,
768
+ "loss": 0.5953,
769
+ "step": 120
770
+ },
771
+ {
772
+ "epoch": 1.4,
773
+ "eval_loss": 0.5730018615722656,
774
+ "eval_runtime": 2.202,
775
+ "eval_samples_per_second": 13.624,
776
+ "eval_steps_per_second": 6.812,
777
+ "step": 120
778
+ },
779
+ {
780
+ "epoch": 1.41,
781
+ "learning_rate": 0.00010676550198477293,
782
+ "loss": 0.6097,
783
+ "step": 121
784
+ },
785
+ {
786
+ "epoch": 1.42,
787
+ "learning_rate": 0.00010541389085854176,
788
+ "loss": 0.6257,
789
+ "step": 122
790
+ },
791
+ {
792
+ "epoch": 1.43,
793
+ "learning_rate": 0.00010406128701262128,
794
+ "loss": 0.6732,
795
+ "step": 123
796
+ },
797
+ {
798
+ "epoch": 1.44,
799
+ "learning_rate": 0.00010270793846761347,
800
+ "loss": 0.6086,
801
+ "step": 124
802
+ },
803
+ {
804
+ "epoch": 1.45,
805
+ "learning_rate": 0.00010135409338067219,
806
+ "loss": 0.6072,
807
+ "step": 125
808
+ },
809
+ {
810
+ "epoch": 1.47,
811
+ "learning_rate": 0.0001,
812
+ "loss": 0.5915,
813
+ "step": 126
814
+ },
815
+ {
816
+ "epoch": 1.48,
817
+ "learning_rate": 9.864590661932783e-05,
818
+ "loss": 0.594,
819
+ "step": 127
820
+ },
821
+ {
822
+ "epoch": 1.49,
823
+ "learning_rate": 9.729206153238657e-05,
824
+ "loss": 0.615,
825
+ "step": 128
826
+ },
827
+ {
828
+ "epoch": 1.5,
829
+ "learning_rate": 9.59387129873787e-05,
830
+ "loss": 0.6108,
831
+ "step": 129
832
+ },
833
+ {
834
+ "epoch": 1.51,
835
+ "learning_rate": 9.458610914145826e-05,
836
+ "loss": 0.6013,
837
+ "step": 130
838
+ },
839
+ {
840
+ "epoch": 1.52,
841
+ "learning_rate": 9.323449801522709e-05,
842
+ "loss": 0.6717,
843
+ "step": 131
844
+ },
845
+ {
846
+ "epoch": 1.53,
847
+ "learning_rate": 9.18841274472569e-05,
848
+ "loss": 0.5898,
849
+ "step": 132
850
+ },
851
+ {
852
+ "epoch": 1.55,
853
+ "learning_rate": 9.05352450486439e-05,
854
+ "loss": 0.5968,
855
+ "step": 133
856
+ },
857
+ {
858
+ "epoch": 1.56,
859
+ "learning_rate": 8.918809815760585e-05,
860
+ "loss": 0.591,
861
+ "step": 134
862
+ },
863
+ {
864
+ "epoch": 1.57,
865
+ "learning_rate": 8.78429337941285e-05,
866
+ "loss": 0.6031,
867
+ "step": 135
868
+ },
869
+ {
870
+ "epoch": 1.58,
871
+ "learning_rate": 8.649999861467099e-05,
872
+ "loss": 0.6373,
873
+ "step": 136
874
+ },
875
+ {
876
+ "epoch": 1.59,
877
+ "learning_rate": 8.515953886693739e-05,
878
+ "loss": 0.596,
879
+ "step": 137
880
+ },
881
+ {
882
+ "epoch": 1.6,
883
+ "learning_rate": 8.382180034472353e-05,
884
+ "loss": 0.6255,
885
+ "step": 138
886
+ },
887
+ {
888
+ "epoch": 1.62,
889
+ "learning_rate": 8.248702834284693e-05,
890
+ "loss": 0.6816,
891
+ "step": 139
892
+ },
893
+ {
894
+ "epoch": 1.63,
895
+ "learning_rate": 8.115546761216822e-05,
896
+ "loss": 0.6179,
897
+ "step": 140
898
+ },
899
+ {
900
+ "epoch": 1.63,
901
+ "eval_loss": 0.5733774900436401,
902
+ "eval_runtime": 2.1935,
903
+ "eval_samples_per_second": 13.677,
904
+ "eval_steps_per_second": 6.838,
905
+ "step": 140
906
+ },
907
+ {
908
+ "epoch": 1.64,
909
+ "learning_rate": 7.982736231471224e-05,
910
+ "loss": 0.6495,
911
+ "step": 141
912
+ },
913
+ {
914
+ "epoch": 1.65,
915
+ "learning_rate": 7.85029559788976e-05,
916
+ "loss": 0.5634,
917
+ "step": 142
918
+ },
919
+ {
920
+ "epoch": 1.66,
921
+ "learning_rate": 7.718249145488142e-05,
922
+ "loss": 0.6272,
923
+ "step": 143
924
+ },
925
+ {
926
+ "epoch": 1.67,
927
+ "learning_rate": 7.586621087002945e-05,
928
+ "loss": 0.6017,
929
+ "step": 144
930
+ },
931
+ {
932
+ "epoch": 1.69,
933
+ "learning_rate": 7.455435558451823e-05,
934
+ "loss": 0.6036,
935
+ "step": 145
936
+ },
937
+ {
938
+ "epoch": 1.7,
939
+ "learning_rate": 7.324716614707793e-05,
940
+ "loss": 0.6083,
941
+ "step": 146
942
+ },
943
+ {
944
+ "epoch": 1.71,
945
+ "learning_rate": 7.194488225088417e-05,
946
+ "loss": 0.6059,
947
+ "step": 147
948
+ },
949
+ {
950
+ "epoch": 1.72,
951
+ "learning_rate": 7.064774268960653e-05,
952
+ "loss": 0.634,
953
+ "step": 148
954
+ },
955
+ {
956
+ "epoch": 1.73,
957
+ "learning_rate": 6.93559853136221e-05,
958
+ "loss": 0.6108,
959
+ "step": 149
960
+ },
961
+ {
962
+ "epoch": 1.74,
963
+ "learning_rate": 6.806984698640202e-05,
964
+ "loss": 0.5924,
965
+ "step": 150
966
+ },
967
+ {
968
+ "epoch": 1.76,
969
+ "learning_rate": 6.678956354107882e-05,
970
+ "loss": 0.62,
971
+ "step": 151
972
+ },
973
+ {
974
+ "epoch": 1.77,
975
+ "learning_rate": 6.551536973720298e-05,
976
+ "loss": 0.6059,
977
+ "step": 152
978
+ },
979
+ {
980
+ "epoch": 1.78,
981
+ "learning_rate": 6.4247499217696e-05,
982
+ "loss": 0.6579,
983
+ "step": 153
984
+ },
985
+ {
986
+ "epoch": 1.79,
987
+ "learning_rate": 6.298618446600856e-05,
988
+ "loss": 0.6381,
989
+ "step": 154
990
+ },
991
+ {
992
+ "epoch": 1.8,
993
+ "learning_rate": 6.173165676349103e-05,
994
+ "loss": 0.6084,
995
+ "step": 155
996
+ },
997
+ {
998
+ "epoch": 1.81,
999
+ "learning_rate": 6.048414614698448e-05,
1000
+ "loss": 0.6106,
1001
+ "step": 156
1002
+ },
1003
+ {
1004
+ "epoch": 1.83,
1005
+ "learning_rate": 5.924388136663992e-05,
1006
+ "loss": 0.6583,
1007
+ "step": 157
1008
+ },
1009
+ {
1010
+ "epoch": 1.84,
1011
+ "learning_rate": 5.801108984397354e-05,
1012
+ "loss": 0.6216,
1013
+ "step": 158
1014
+ },
1015
+ {
1016
+ "epoch": 1.85,
1017
+ "learning_rate": 5.6785997630165435e-05,
1018
+ "loss": 0.6261,
1019
+ "step": 159
1020
+ },
1021
+ {
1022
+ "epoch": 1.86,
1023
+ "learning_rate": 5.5568829364609664e-05,
1024
+ "loss": 0.6035,
1025
+ "step": 160
1026
+ },
1027
+ {
1028
+ "epoch": 1.86,
1029
+ "eval_loss": 0.5652638673782349,
1030
+ "eval_runtime": 2.1857,
1031
+ "eval_samples_per_second": 13.726,
1032
+ "eval_steps_per_second": 6.863,
1033
+ "step": 160
1034
+ },
1035
+ {
1036
+ "epoch": 1.87,
1037
+ "learning_rate": 5.435980823372311e-05,
1038
+ "loss": 0.6083,
1039
+ "step": 161
1040
+ },
1041
+ {
1042
+ "epoch": 1.88,
1043
+ "learning_rate": 5.3159155930021e-05,
1044
+ "loss": 0.627,
1045
+ "step": 162
1046
+ },
1047
+ {
1048
+ "epoch": 1.9,
1049
+ "learning_rate": 5.196709261146606e-05,
1050
+ "loss": 0.6006,
1051
+ "step": 163
1052
+ },
1053
+ {
1054
+ "epoch": 1.91,
1055
+ "learning_rate": 5.078383686109926e-05,
1056
+ "loss": 0.5821,
1057
+ "step": 164
1058
+ },
1059
+ {
1060
+ "epoch": 1.92,
1061
+ "learning_rate": 4.9609605646959226e-05,
1062
+ "loss": 0.6239,
1063
+ "step": 165
1064
+ },
1065
+ {
1066
+ "epoch": 1.93,
1067
+ "learning_rate": 4.844461428229782e-05,
1068
+ "loss": 0.5951,
1069
+ "step": 166
1070
+ },
1071
+ {
1072
+ "epoch": 1.94,
1073
+ "learning_rate": 4.728907638609925e-05,
1074
+ "loss": 0.6376,
1075
+ "step": 167
1076
+ },
1077
+ {
1078
+ "epoch": 1.95,
1079
+ "learning_rate": 4.614320384390959e-05,
1080
+ "loss": 0.6346,
1081
+ "step": 168
1082
+ },
1083
+ {
1084
+ "epoch": 1.97,
1085
+ "learning_rate": 4.500720676898452e-05,
1086
+ "loss": 0.6373,
1087
+ "step": 169
1088
+ },
1089
+ {
1090
+ "epoch": 1.98,
1091
+ "learning_rate": 4.388129346376178e-05,
1092
+ "loss": 0.5886,
1093
+ "step": 170
1094
+ },
1095
+ {
1096
+ "epoch": 1.99,
1097
+ "learning_rate": 4.276567038166563e-05,
1098
+ "loss": 0.6058,
1099
+ "step": 171
1100
+ },
1101
+ {
1102
+ "epoch": 2.0,
1103
+ "learning_rate": 4.16605420892506e-05,
1104
+ "loss": 0.6359,
1105
+ "step": 172
1106
+ },
1107
+ {
1108
+ "epoch": 2.01,
1109
+ "learning_rate": 4.0566111228691064e-05,
1110
+ "loss": 0.601,
1111
+ "step": 173
1112
+ },
1113
+ {
1114
+ "epoch": 2.02,
1115
+ "learning_rate": 3.948257848062351e-05,
1116
+ "loss": 0.5853,
1117
+ "step": 174
1118
+ },
1119
+ {
1120
+ "epoch": 2.03,
1121
+ "learning_rate": 3.841014252734896e-05,
1122
+ "loss": 0.5868,
1123
+ "step": 175
1124
+ },
1125
+ {
1126
+ "epoch": 2.05,
1127
+ "learning_rate": 3.734900001640135e-05,
1128
+ "loss": 0.582,
1129
+ "step": 176
1130
+ },
1131
+ {
1132
+ "epoch": 2.06,
1133
+ "learning_rate": 3.629934552448925e-05,
1134
+ "loss": 0.5974,
1135
+ "step": 177
1136
+ },
1137
+ {
1138
+ "epoch": 2.07,
1139
+ "learning_rate": 3.5261371521817244e-05,
1140
+ "loss": 0.6215,
1141
+ "step": 178
1142
+ },
1143
+ {
1144
+ "epoch": 2.08,
1145
+ "learning_rate": 3.423526833679355e-05,
1146
+ "loss": 0.6196,
1147
+ "step": 179
1148
+ },
1149
+ {
1150
+ "epoch": 2.09,
1151
+ "learning_rate": 3.322122412113047e-05,
1152
+ "loss": 0.5434,
1153
+ "step": 180
1154
+ },
1155
+ {
1156
+ "epoch": 2.09,
1157
+ "eval_loss": 0.5678020715713501,
1158
+ "eval_runtime": 2.1856,
1159
+ "eval_samples_per_second": 13.726,
1160
+ "eval_steps_per_second": 6.863,
1161
+ "step": 180
1162
+ },
1163
+ {
1164
+ "epoch": 2.1,
1165
+ "learning_rate": 3.2219424815343735e-05,
1166
+ "loss": 0.621,
1167
+ "step": 181
1168
+ },
1169
+ {
1170
+ "epoch": 2.12,
1171
+ "learning_rate": 3.123005411465766e-05,
1172
+ "loss": 0.5455,
1173
+ "step": 182
1174
+ },
1175
+ {
1176
+ "epoch": 2.13,
1177
+ "learning_rate": 3.0253293435321793e-05,
1178
+ "loss": 0.5503,
1179
+ "step": 183
1180
+ },
1181
+ {
1182
+ "epoch": 2.14,
1183
+ "learning_rate": 2.9289321881345254e-05,
1184
+ "loss": 0.578,
1185
+ "step": 184
1186
+ },
1187
+ {
1188
+ "epoch": 2.15,
1189
+ "learning_rate": 2.8338316211655536e-05,
1190
+ "loss": 0.5664,
1191
+ "step": 185
1192
+ },
1193
+ {
1194
+ "epoch": 2.16,
1195
+ "learning_rate": 2.7400450807686938e-05,
1196
+ "loss": 0.5854,
1197
+ "step": 186
1198
+ },
1199
+ {
1200
+ "epoch": 2.17,
1201
+ "learning_rate": 2.647589764140499e-05,
1202
+ "loss": 0.593,
1203
+ "step": 187
1204
+ },
1205
+ {
1206
+ "epoch": 2.19,
1207
+ "learning_rate": 2.5564826243772966e-05,
1208
+ "loss": 0.5993,
1209
+ "step": 188
1210
+ },
1211
+ {
1212
+ "epoch": 2.2,
1213
+ "learning_rate": 2.4667403673665623e-05,
1214
+ "loss": 0.6146,
1215
+ "step": 189
1216
+ },
1217
+ {
1218
+ "epoch": 2.21,
1219
+ "learning_rate": 2.3783794487236365e-05,
1220
+ "loss": 0.5969,
1221
+ "step": 190
1222
+ },
1223
+ {
1224
+ "epoch": 2.22,
1225
+ "learning_rate": 2.2914160707743538e-05,
1226
+ "loss": 0.6258,
1227
+ "step": 191
1228
+ },
1229
+ {
1230
+ "epoch": 2.23,
1231
+ "learning_rate": 2.205866179584084e-05,
1232
+ "loss": 0.592,
1233
+ "step": 192
1234
+ },
1235
+ {
1236
+ "epoch": 2.24,
1237
+ "learning_rate": 2.121745462033784e-05,
1238
+ "loss": 0.5576,
1239
+ "step": 193
1240
+ },
1241
+ {
1242
+ "epoch": 2.26,
1243
+ "learning_rate": 2.0390693429435627e-05,
1244
+ "loss": 0.6122,
1245
+ "step": 194
1246
+ },
1247
+ {
1248
+ "epoch": 2.27,
1249
+ "learning_rate": 1.957852982244309e-05,
1250
+ "loss": 0.5836,
1251
+ "step": 195
1252
+ },
1253
+ {
1254
+ "epoch": 2.28,
1255
+ "learning_rate": 1.87811127219789e-05,
1256
+ "loss": 0.6006,
1257
+ "step": 196
1258
+ },
1259
+ {
1260
+ "epoch": 2.29,
1261
+ "learning_rate": 1.7998588346664115e-05,
1262
+ "loss": 0.5786,
1263
+ "step": 197
1264
+ },
1265
+ {
1266
+ "epoch": 2.3,
1267
+ "learning_rate": 1.7231100184310956e-05,
1268
+ "loss": 0.599,
1269
+ "step": 198
1270
+ },
1271
+ {
1272
+ "epoch": 2.31,
1273
+ "learning_rate": 1.6478788965611993e-05,
1274
+ "loss": 0.5833,
1275
+ "step": 199
1276
+ },
1277
+ {
1278
+ "epoch": 2.33,
1279
+ "learning_rate": 1.5741792638335095e-05,
1280
+ "loss": 0.6036,
1281
+ "step": 200
1282
+ },
1283
+ {
1284
+ "epoch": 2.33,
1285
+ "eval_loss": 0.5672248005867004,
1286
+ "eval_runtime": 2.1898,
1287
+ "eval_samples_per_second": 13.7,
1288
+ "eval_steps_per_second": 6.85,
1289
+ "step": 200
1290
+ },
1291
+ {
1292
+ "epoch": 2.34,
1293
+ "learning_rate": 1.502024634202851e-05,
1294
+ "loss": 0.5703,
1295
+ "step": 201
1296
+ },
1297
+ {
1298
+ "epoch": 2.35,
1299
+ "learning_rate": 1.4314282383241096e-05,
1300
+ "loss": 0.5941,
1301
+ "step": 202
1302
+ },
1303
+ {
1304
+ "epoch": 2.36,
1305
+ "learning_rate": 1.3624030211261685e-05,
1306
+ "loss": 0.5736,
1307
+ "step": 203
1308
+ },
1309
+ {
1310
+ "epoch": 2.37,
1311
+ "learning_rate": 1.2949616394382802e-05,
1312
+ "loss": 0.6027,
1313
+ "step": 204
1314
+ },
1315
+ {
1316
+ "epoch": 2.38,
1317
+ "learning_rate": 1.2291164596692305e-05,
1318
+ "loss": 0.608,
1319
+ "step": 205
1320
+ },
1321
+ {
1322
+ "epoch": 2.4,
1323
+ "learning_rate": 1.1648795555397719e-05,
1324
+ "loss": 0.5645,
1325
+ "step": 206
1326
+ },
1327
+ {
1328
+ "epoch": 2.41,
1329
+ "learning_rate": 1.1022627058687163e-05,
1330
+ "loss": 0.5655,
1331
+ "step": 207
1332
+ },
1333
+ {
1334
+ "epoch": 2.42,
1335
+ "learning_rate": 1.0412773924131203e-05,
1336
+ "loss": 0.5954,
1337
+ "step": 208
1338
+ },
1339
+ {
1340
+ "epoch": 2.43,
1341
+ "learning_rate": 9.819347977629202e-06,
1342
+ "loss": 0.5703,
1343
+ "step": 209
1344
+ },
1345
+ {
1346
+ "epoch": 2.44,
1347
+ "learning_rate": 9.242458032904311e-06,
1348
+ "loss": 0.5942,
1349
+ "step": 210
1350
+ },
1351
+ {
1352
+ "epoch": 2.45,
1353
+ "learning_rate": 8.682209871550884e-06,
1354
+ "loss": 0.5895,
1355
+ "step": 211
1356
+ },
1357
+ {
1358
+ "epoch": 2.47,
1359
+ "learning_rate": 8.138706223637827e-06,
1360
+ "loss": 0.583,
1361
+ "step": 212
1362
+ },
1363
+ {
1364
+ "epoch": 2.48,
1365
+ "learning_rate": 7.612046748871327e-06,
1366
+ "loss": 0.6183,
1367
+ "step": 213
1368
+ },
1369
+ {
1370
+ "epoch": 2.49,
1371
+ "learning_rate": 7.102328018320858e-06,
1372
+ "loss": 0.577,
1373
+ "step": 214
1374
+ },
1375
+ {
1376
+ "epoch": 2.5,
1377
+ "learning_rate": 6.609643496711349e-06,
1378
+ "loss": 0.5897,
1379
+ "step": 215
1380
+ },
1381
+ {
1382
+ "epoch": 2.51,
1383
+ "learning_rate": 6.13408352528495e-06,
1384
+ "loss": 0.5862,
1385
+ "step": 216
1386
+ },
1387
+ {
1388
+ "epoch": 2.52,
1389
+ "learning_rate": 5.6757353052356964e-06,
1390
+ "loss": 0.6058,
1391
+ "step": 217
1392
+ },
1393
+ {
1394
+ "epoch": 2.53,
1395
+ "learning_rate": 5.2346828817197655e-06,
1396
+ "loss": 0.5563,
1397
+ "step": 218
1398
+ },
1399
+ {
1400
+ "epoch": 2.55,
1401
+ "learning_rate": 4.811007128444445e-06,
1402
+ "loss": 0.5681,
1403
+ "step": 219
1404
+ },
1405
+ {
1406
+ "epoch": 2.56,
1407
+ "learning_rate": 4.404785732838846e-06,
1408
+ "loss": 0.5766,
1409
+ "step": 220
1410
+ },
1411
+ {
1412
+ "epoch": 2.56,
1413
+ "eval_loss": 0.5669747591018677,
1414
+ "eval_runtime": 2.1807,
1415
+ "eval_samples_per_second": 13.757,
1416
+ "eval_steps_per_second": 6.878,
1417
+ "step": 220
1418
+ },
1419
+ {
1420
+ "epoch": 2.57,
1421
+ "learning_rate": 4.016093181808623e-06,
1422
+ "loss": 0.5869,
1423
+ "step": 221
1424
+ },
1425
+ {
1426
+ "epoch": 2.58,
1427
+ "learning_rate": 3.6450007480777093e-06,
1428
+ "loss": 0.584,
1429
+ "step": 222
1430
+ },
1431
+ {
1432
+ "epoch": 2.59,
1433
+ "learning_rate": 3.2915764771193292e-06,
1434
+ "loss": 0.6152,
1435
+ "step": 223
1436
+ },
1437
+ {
1438
+ "epoch": 2.6,
1439
+ "learning_rate": 2.9558851746788517e-06,
1440
+ "loss": 0.5596,
1441
+ "step": 224
1442
+ },
1443
+ {
1444
+ "epoch": 2.62,
1445
+ "learning_rate": 2.6379883948907e-06,
1446
+ "loss": 0.55,
1447
+ "step": 225
1448
+ },
1449
+ {
1450
+ "epoch": 2.63,
1451
+ "learning_rate": 2.3379444289913342e-06,
1452
+ "loss": 0.5853,
1453
+ "step": 226
1454
+ },
1455
+ {
1456
+ "epoch": 2.64,
1457
+ "learning_rate": 2.0558082946308232e-06,
1458
+ "loss": 0.6064,
1459
+ "step": 227
1460
+ },
1461
+ {
1462
+ "epoch": 2.65,
1463
+ "learning_rate": 1.7916317257844039e-06,
1464
+ "loss": 0.6225,
1465
+ "step": 228
1466
+ },
1467
+ {
1468
+ "epoch": 2.66,
1469
+ "learning_rate": 1.545463163266303e-06,
1470
+ "loss": 0.6052,
1471
+ "step": 229
1472
+ },
1473
+ {
1474
+ "epoch": 2.67,
1475
+ "learning_rate": 1.317347745847386e-06,
1476
+ "loss": 0.5799,
1477
+ "step": 230
1478
+ },
1479
+ {
1480
+ "epoch": 2.69,
1481
+ "learning_rate": 1.107327301978245e-06,
1482
+ "loss": 0.572,
1483
+ "step": 231
1484
+ },
1485
+ {
1486
+ "epoch": 2.7,
1487
+ "learning_rate": 9.154403421193225e-07,
1488
+ "loss": 0.574,
1489
+ "step": 232
1490
+ },
1491
+ {
1492
+ "epoch": 2.71,
1493
+ "learning_rate": 7.417220516794499e-07,
1494
+ "loss": 0.5924,
1495
+ "step": 233
1496
+ },
1497
+ {
1498
+ "epoch": 2.72,
1499
+ "learning_rate": 5.862042845640403e-07,
1500
+ "loss": 0.5591,
1501
+ "step": 234
1502
+ },
1503
+ {
1504
+ "epoch": 2.73,
1505
+ "learning_rate": 4.489155573341841e-07,
1506
+ "loss": 0.5707,
1507
+ "step": 235
1508
+ },
1509
+ {
1510
+ "epoch": 2.74,
1511
+ "learning_rate": 3.298810439777311e-07,
1512
+ "loss": 0.5837,
1513
+ "step": 236
1514
+ },
1515
+ {
1516
+ "epoch": 2.76,
1517
+ "learning_rate": 2.2912257129320547e-07,
1518
+ "loss": 0.6152,
1519
+ "step": 237
1520
+ },
1521
+ {
1522
+ "epoch": 2.77,
1523
+ "learning_rate": 1.4665861488761813e-07,
1524
+ "loss": 0.5879,
1525
+ "step": 238
1526
+ },
1527
+ {
1528
+ "epoch": 2.78,
1529
+ "learning_rate": 8.250429578855467e-08,
1530
+ "loss": 0.6049,
1531
+ "step": 239
1532
+ },
1533
+ {
1534
+ "epoch": 2.79,
1535
+ "learning_rate": 3.667137767160433e-08,
1536
+ "loss": 0.5795,
1537
+ "step": 240
1538
+ },
1539
+ {
1540
+ "epoch": 2.79,
1541
+ "eval_loss": 0.5656673908233643,
1542
+ "eval_runtime": 2.2018,
1543
+ "eval_samples_per_second": 13.625,
1544
+ "eval_steps_per_second": 6.813,
1545
+ "step": 240
1546
+ },
1547
+ {
1548
+ "epoch": 2.8,
1549
+ "learning_rate": 9.168264703285356e-09,
1550
+ "loss": 0.5562,
1551
+ "step": 241
1552
+ },
1553
+ {
1554
+ "epoch": 2.81,
1555
+ "learning_rate": 0.0,
1556
+ "loss": 0.5535,
1557
+ "step": 242
1558
+ },
1559
+ {
1560
+ "epoch": 2.83,
1561
+ "learning_rate": 9.168264703285356e-09,
1562
+ "loss": 0.5643,
1563
+ "step": 243
1564
+ },
1565
+ {
1566
+ "epoch": 2.84,
1567
+ "learning_rate": 3.667137767160433e-08,
1568
+ "loss": 0.5899,
1569
+ "step": 244
1570
+ },
1571
+ {
1572
+ "epoch": 2.85,
1573
+ "learning_rate": 8.250429578855467e-08,
1574
+ "loss": 0.6018,
1575
+ "step": 245
1576
+ },
1577
+ {
1578
+ "epoch": 2.86,
1579
+ "learning_rate": 1.46658614887607e-07,
1580
+ "loss": 0.6263,
1581
+ "step": 246
1582
+ },
1583
+ {
1584
+ "epoch": 2.87,
1585
+ "learning_rate": 2.2912257129320547e-07,
1586
+ "loss": 0.6012,
1587
+ "step": 247
1588
+ },
1589
+ {
1590
+ "epoch": 2.88,
1591
+ "learning_rate": 3.298810439777311e-07,
1592
+ "loss": 0.5891,
1593
+ "step": 248
1594
+ },
1595
+ {
1596
+ "epoch": 2.9,
1597
+ "learning_rate": 4.489155573341841e-07,
1598
+ "loss": 0.5751,
1599
+ "step": 249
1600
+ },
1601
+ {
1602
+ "epoch": 2.91,
1603
+ "learning_rate": 5.862042845640403e-07,
1604
+ "loss": 0.5787,
1605
+ "step": 250
1606
+ },
1607
+ {
1608
+ "epoch": 2.92,
1609
+ "learning_rate": 7.417220516794499e-07,
1610
+ "loss": 0.5652,
1611
+ "step": 251
1612
+ },
1613
+ {
1614
+ "epoch": 2.93,
1615
+ "learning_rate": 9.154403421193225e-07,
1616
+ "loss": 0.6027,
1617
+ "step": 252
1618
+ },
1619
+ {
1620
+ "epoch": 2.94,
1621
+ "learning_rate": 1.1073273019782337e-06,
1622
+ "loss": 0.5621,
1623
+ "step": 253
1624
+ },
1625
+ {
1626
+ "epoch": 2.95,
1627
+ "learning_rate": 1.3173477458473749e-06,
1628
+ "loss": 0.6019,
1629
+ "step": 254
1630
+ },
1631
+ {
1632
+ "epoch": 2.97,
1633
+ "learning_rate": 1.545463163266303e-06,
1634
+ "loss": 0.5759,
1635
+ "step": 255
1636
+ },
1637
+ {
1638
+ "epoch": 2.98,
1639
+ "learning_rate": 1.7916317257843929e-06,
1640
+ "loss": 0.58,
1641
+ "step": 256
1642
+ },
1643
+ {
1644
+ "epoch": 2.99,
1645
+ "learning_rate": 2.0558082946308232e-06,
1646
+ "loss": 0.5375,
1647
+ "step": 257
1648
+ },
1649
+ {
1650
+ "epoch": 3.0,
1651
+ "learning_rate": 2.3379444289913232e-06,
1652
+ "loss": 0.5893,
1653
+ "step": 258
1654
+ }
1655
+ ],
1656
+ "logging_steps": 1,
1657
+ "max_steps": 258,
1658
+ "num_train_epochs": 3,
1659
+ "save_steps": 500,
1660
+ "total_flos": 3.195702207216353e+17,
1661
+ "trial_name": null,
1662
+ "trial_params": null
1663
+ }
checkpoints/checkpoint-258/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:50c9227896cee3730521610abbbf3be77cf9b1863b67948e92913b6b096f2c5c
3
+ size 4219