tyzhu commited on
Commit
e3b00c0
1 Parent(s): 3f4dafd

Training in progress, epoch 22, checkpoint

Browse files
checkpoint-25052/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.5.0
checkpoint-25052/adapter_config.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": false,
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": 16,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "k_proj",
19
+ "v_proj",
20
+ "down_proj",
21
+ "gate_proj",
22
+ "up_proj"
23
+ ],
24
+ "task_type": "CAUSAL_LM"
25
+ }
checkpoint-25052/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3a11faaeb715e991e49928321ac4e1bd7c63fc6098af21f408257e3f00e459e2
3
+ size 143269386
checkpoint-25052/added_tokens.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "</s>": 2,
3
+ "<s>": 1,
4
+ "<unk>": 0
5
+ }
checkpoint-25052/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ddfdb2084328d1bffee5b087d17f9ac0e4864f425d42ff9d06d70607a054c25e
3
+ size 286585234
checkpoint-25052/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6f7e687ecc4a541e7d134807163e0c6278bc78819812c1e23371d4ecd14d2e42
3
+ size 15024
checkpoint-25052/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:684cff78665f9457a41522f832a47d288a5ce185ab34016b5f5c01347899c89e
3
+ size 15024
checkpoint-25052/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:595feebc14a164c1d16d47e5e8a58fb4e57e3b11c2ef421d75a56fc5259a72c7
3
+ size 15024
checkpoint-25052/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58d1656883ed76ca4bc1e0bbf70d4bd9ff739e93dd3f29dfe3de95acc59e82a5
3
+ size 15024
checkpoint-25052/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aecafe3190946daffb466f91530c1e161ded649fb189bf7003872d607493d9ad
3
+ size 1064
checkpoint-25052/special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "pad_token": "</s>",
5
+ "unk_token": "<unk>"
6
+ }
checkpoint-25052/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-25052/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-25052/tokenizer_config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<unk>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<s>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ }
27
+ },
28
+ "additional_special_tokens": [],
29
+ "bos_token": "<s>",
30
+ "clean_up_tokenization_spaces": false,
31
+ "eos_token": "</s>",
32
+ "legacy": false,
33
+ "model_max_length": 1000000000000000019884624838656,
34
+ "pad_token": "</s>",
35
+ "padding_side": "left",
36
+ "sp_model_kwargs": {},
37
+ "tokenizer_class": "LlamaTokenizer",
38
+ "unk_token": "<unk>",
39
+ "use_default_system_prompt": true
40
+ }
checkpoint-25052/trainer_state.json ADDED
@@ -0,0 +1,1864 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 22.99931145283452,
5
+ "eval_steps": 500,
6
+ "global_step": 25052,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.09,
13
+ "learning_rate": 3e-05,
14
+ "loss": 1.5477,
15
+ "step": 100
16
+ },
17
+ {
18
+ "epoch": 0.18,
19
+ "learning_rate": 3e-05,
20
+ "loss": 1.3248,
21
+ "step": 200
22
+ },
23
+ {
24
+ "epoch": 0.28,
25
+ "learning_rate": 3e-05,
26
+ "loss": 1.2394,
27
+ "step": 300
28
+ },
29
+ {
30
+ "epoch": 0.37,
31
+ "learning_rate": 3e-05,
32
+ "loss": 1.2549,
33
+ "step": 400
34
+ },
35
+ {
36
+ "epoch": 0.46,
37
+ "learning_rate": 3e-05,
38
+ "loss": 1.2163,
39
+ "step": 500
40
+ },
41
+ {
42
+ "epoch": 0.55,
43
+ "learning_rate": 3e-05,
44
+ "loss": 1.2202,
45
+ "step": 600
46
+ },
47
+ {
48
+ "epoch": 0.64,
49
+ "learning_rate": 3e-05,
50
+ "loss": 1.2099,
51
+ "step": 700
52
+ },
53
+ {
54
+ "epoch": 0.73,
55
+ "learning_rate": 3e-05,
56
+ "loss": 1.2213,
57
+ "step": 800
58
+ },
59
+ {
60
+ "epoch": 0.83,
61
+ "learning_rate": 3e-05,
62
+ "loss": 1.2203,
63
+ "step": 900
64
+ },
65
+ {
66
+ "epoch": 0.92,
67
+ "learning_rate": 3e-05,
68
+ "loss": 1.2166,
69
+ "step": 1000
70
+ },
71
+ {
72
+ "epoch": 1.0,
73
+ "eval_accuracy": 0.5925822784810126,
74
+ "eval_loss": 1.858124017715454,
75
+ "eval_runtime": 4.7843,
76
+ "eval_samples_per_second": 104.509,
77
+ "eval_steps_per_second": 13.168,
78
+ "step": 1089
79
+ },
80
+ {
81
+ "epoch": 1.0,
82
+ "eval_exact_match": 25.0,
83
+ "eval_f1": 33.90013632566266,
84
+ "step": 1089
85
+ },
86
+ {
87
+ "epoch": 1.01,
88
+ "learning_rate": 3e-05,
89
+ "loss": 1.2141,
90
+ "step": 1100
91
+ },
92
+ {
93
+ "epoch": 1.1,
94
+ "learning_rate": 3e-05,
95
+ "loss": 1.2178,
96
+ "step": 1200
97
+ },
98
+ {
99
+ "epoch": 1.19,
100
+ "learning_rate": 3e-05,
101
+ "loss": 1.1986,
102
+ "step": 1300
103
+ },
104
+ {
105
+ "epoch": 1.29,
106
+ "learning_rate": 3e-05,
107
+ "loss": 1.1903,
108
+ "step": 1400
109
+ },
110
+ {
111
+ "epoch": 1.38,
112
+ "learning_rate": 3e-05,
113
+ "loss": 1.1925,
114
+ "step": 1500
115
+ },
116
+ {
117
+ "epoch": 1.47,
118
+ "learning_rate": 3e-05,
119
+ "loss": 1.1954,
120
+ "step": 1600
121
+ },
122
+ {
123
+ "epoch": 1.56,
124
+ "learning_rate": 3e-05,
125
+ "loss": 1.1859,
126
+ "step": 1700
127
+ },
128
+ {
129
+ "epoch": 1.65,
130
+ "learning_rate": 3e-05,
131
+ "loss": 1.1979,
132
+ "step": 1800
133
+ },
134
+ {
135
+ "epoch": 1.74,
136
+ "learning_rate": 3e-05,
137
+ "loss": 1.201,
138
+ "step": 1900
139
+ },
140
+ {
141
+ "epoch": 1.84,
142
+ "learning_rate": 3e-05,
143
+ "loss": 1.1871,
144
+ "step": 2000
145
+ },
146
+ {
147
+ "epoch": 1.93,
148
+ "learning_rate": 3e-05,
149
+ "loss": 1.1949,
150
+ "step": 2100
151
+ },
152
+ {
153
+ "epoch": 2.0,
154
+ "eval_accuracy": 0.5960506329113924,
155
+ "eval_loss": 1.834545373916626,
156
+ "eval_runtime": 4.5511,
157
+ "eval_samples_per_second": 109.863,
158
+ "eval_steps_per_second": 13.843,
159
+ "step": 2178
160
+ },
161
+ {
162
+ "epoch": 2.0,
163
+ "eval_exact_match": 25.2,
164
+ "eval_f1": 34.41373827926461,
165
+ "step": 2178
166
+ },
167
+ {
168
+ "epoch": 2.02,
169
+ "learning_rate": 3e-05,
170
+ "loss": 1.1775,
171
+ "step": 2200
172
+ },
173
+ {
174
+ "epoch": 2.11,
175
+ "learning_rate": 3e-05,
176
+ "loss": 1.1579,
177
+ "step": 2300
178
+ },
179
+ {
180
+ "epoch": 2.2,
181
+ "learning_rate": 3e-05,
182
+ "loss": 1.1678,
183
+ "step": 2400
184
+ },
185
+ {
186
+ "epoch": 2.3,
187
+ "learning_rate": 3e-05,
188
+ "loss": 1.1758,
189
+ "step": 2500
190
+ },
191
+ {
192
+ "epoch": 2.39,
193
+ "learning_rate": 3e-05,
194
+ "loss": 1.1374,
195
+ "step": 2600
196
+ },
197
+ {
198
+ "epoch": 2.48,
199
+ "learning_rate": 3e-05,
200
+ "loss": 1.16,
201
+ "step": 2700
202
+ },
203
+ {
204
+ "epoch": 2.57,
205
+ "learning_rate": 3e-05,
206
+ "loss": 1.1647,
207
+ "step": 2800
208
+ },
209
+ {
210
+ "epoch": 2.66,
211
+ "learning_rate": 3e-05,
212
+ "loss": 1.1579,
213
+ "step": 2900
214
+ },
215
+ {
216
+ "epoch": 2.75,
217
+ "learning_rate": 3e-05,
218
+ "loss": 1.1647,
219
+ "step": 3000
220
+ },
221
+ {
222
+ "epoch": 2.85,
223
+ "learning_rate": 3e-05,
224
+ "loss": 1.1581,
225
+ "step": 3100
226
+ },
227
+ {
228
+ "epoch": 2.94,
229
+ "learning_rate": 3e-05,
230
+ "loss": 1.1542,
231
+ "step": 3200
232
+ },
233
+ {
234
+ "epoch": 3.0,
235
+ "eval_accuracy": 0.596379746835443,
236
+ "eval_loss": 1.8205524682998657,
237
+ "eval_runtime": 4.5724,
238
+ "eval_samples_per_second": 109.352,
239
+ "eval_steps_per_second": 13.778,
240
+ "step": 3267
241
+ },
242
+ {
243
+ "epoch": 3.0,
244
+ "eval_exact_match": 24.6,
245
+ "eval_f1": 34.42903485403487,
246
+ "step": 3267
247
+ },
248
+ {
249
+ "epoch": 3.03,
250
+ "learning_rate": 3e-05,
251
+ "loss": 1.1336,
252
+ "step": 3300
253
+ },
254
+ {
255
+ "epoch": 3.12,
256
+ "learning_rate": 3e-05,
257
+ "loss": 1.1112,
258
+ "step": 3400
259
+ },
260
+ {
261
+ "epoch": 3.21,
262
+ "learning_rate": 3e-05,
263
+ "loss": 1.123,
264
+ "step": 3500
265
+ },
266
+ {
267
+ "epoch": 3.31,
268
+ "learning_rate": 3e-05,
269
+ "loss": 1.1278,
270
+ "step": 3600
271
+ },
272
+ {
273
+ "epoch": 3.4,
274
+ "learning_rate": 3e-05,
275
+ "loss": 1.1226,
276
+ "step": 3700
277
+ },
278
+ {
279
+ "epoch": 3.49,
280
+ "learning_rate": 3e-05,
281
+ "loss": 1.1132,
282
+ "step": 3800
283
+ },
284
+ {
285
+ "epoch": 3.58,
286
+ "learning_rate": 3e-05,
287
+ "loss": 1.1075,
288
+ "step": 3900
289
+ },
290
+ {
291
+ "epoch": 3.67,
292
+ "learning_rate": 3e-05,
293
+ "loss": 1.128,
294
+ "step": 4000
295
+ },
296
+ {
297
+ "epoch": 3.76,
298
+ "learning_rate": 3e-05,
299
+ "loss": 1.1182,
300
+ "step": 4100
301
+ },
302
+ {
303
+ "epoch": 3.86,
304
+ "learning_rate": 3e-05,
305
+ "loss": 1.1288,
306
+ "step": 4200
307
+ },
308
+ {
309
+ "epoch": 3.95,
310
+ "learning_rate": 3e-05,
311
+ "loss": 1.1236,
312
+ "step": 4300
313
+ },
314
+ {
315
+ "epoch": 4.0,
316
+ "eval_accuracy": 0.600379746835443,
317
+ "eval_loss": 1.822967290878296,
318
+ "eval_runtime": 4.5641,
319
+ "eval_samples_per_second": 109.55,
320
+ "eval_steps_per_second": 13.803,
321
+ "step": 4357
322
+ },
323
+ {
324
+ "epoch": 4.0,
325
+ "eval_exact_match": 23.0,
326
+ "eval_f1": 32.55568709068711,
327
+ "step": 4357
328
+ },
329
+ {
330
+ "epoch": 4.04,
331
+ "learning_rate": 3e-05,
332
+ "loss": 1.1034,
333
+ "step": 4400
334
+ },
335
+ {
336
+ "epoch": 4.13,
337
+ "learning_rate": 3e-05,
338
+ "loss": 1.0822,
339
+ "step": 4500
340
+ },
341
+ {
342
+ "epoch": 4.22,
343
+ "learning_rate": 3e-05,
344
+ "loss": 1.0869,
345
+ "step": 4600
346
+ },
347
+ {
348
+ "epoch": 4.31,
349
+ "learning_rate": 3e-05,
350
+ "loss": 1.0813,
351
+ "step": 4700
352
+ },
353
+ {
354
+ "epoch": 4.41,
355
+ "learning_rate": 3e-05,
356
+ "loss": 1.081,
357
+ "step": 4800
358
+ },
359
+ {
360
+ "epoch": 4.5,
361
+ "learning_rate": 3e-05,
362
+ "loss": 1.1065,
363
+ "step": 4900
364
+ },
365
+ {
366
+ "epoch": 4.59,
367
+ "learning_rate": 3e-05,
368
+ "loss": 1.1018,
369
+ "step": 5000
370
+ },
371
+ {
372
+ "epoch": 4.68,
373
+ "learning_rate": 3e-05,
374
+ "loss": 1.0737,
375
+ "step": 5100
376
+ },
377
+ {
378
+ "epoch": 4.77,
379
+ "learning_rate": 3e-05,
380
+ "loss": 1.0823,
381
+ "step": 5200
382
+ },
383
+ {
384
+ "epoch": 4.87,
385
+ "learning_rate": 3e-05,
386
+ "loss": 1.0916,
387
+ "step": 5300
388
+ },
389
+ {
390
+ "epoch": 4.96,
391
+ "learning_rate": 3e-05,
392
+ "loss": 1.0989,
393
+ "step": 5400
394
+ },
395
+ {
396
+ "epoch": 5.0,
397
+ "eval_accuracy": 0.6035696202531645,
398
+ "eval_loss": 1.8370839357376099,
399
+ "eval_runtime": 4.3939,
400
+ "eval_samples_per_second": 113.795,
401
+ "eval_steps_per_second": 14.338,
402
+ "step": 5446
403
+ },
404
+ {
405
+ "epoch": 5.0,
406
+ "eval_exact_match": 23.8,
407
+ "eval_f1": 33.387894882894905,
408
+ "step": 5446
409
+ },
410
+ {
411
+ "epoch": 5.05,
412
+ "learning_rate": 3e-05,
413
+ "loss": 1.0542,
414
+ "step": 5500
415
+ },
416
+ {
417
+ "epoch": 5.14,
418
+ "learning_rate": 3e-05,
419
+ "loss": 1.0188,
420
+ "step": 5600
421
+ },
422
+ {
423
+ "epoch": 5.23,
424
+ "learning_rate": 3e-05,
425
+ "loss": 1.0364,
426
+ "step": 5700
427
+ },
428
+ {
429
+ "epoch": 5.32,
430
+ "learning_rate": 3e-05,
431
+ "loss": 1.0497,
432
+ "step": 5800
433
+ },
434
+ {
435
+ "epoch": 5.42,
436
+ "learning_rate": 3e-05,
437
+ "loss": 1.0561,
438
+ "step": 5900
439
+ },
440
+ {
441
+ "epoch": 5.51,
442
+ "learning_rate": 3e-05,
443
+ "loss": 1.0538,
444
+ "step": 6000
445
+ },
446
+ {
447
+ "epoch": 5.6,
448
+ "learning_rate": 3e-05,
449
+ "loss": 1.0574,
450
+ "step": 6100
451
+ },
452
+ {
453
+ "epoch": 5.69,
454
+ "learning_rate": 3e-05,
455
+ "loss": 1.0468,
456
+ "step": 6200
457
+ },
458
+ {
459
+ "epoch": 5.78,
460
+ "learning_rate": 3e-05,
461
+ "loss": 1.0524,
462
+ "step": 6300
463
+ },
464
+ {
465
+ "epoch": 5.88,
466
+ "learning_rate": 3e-05,
467
+ "loss": 1.0637,
468
+ "step": 6400
469
+ },
470
+ {
471
+ "epoch": 5.97,
472
+ "learning_rate": 3e-05,
473
+ "loss": 1.0543,
474
+ "step": 6500
475
+ },
476
+ {
477
+ "epoch": 6.0,
478
+ "eval_accuracy": 0.6021518987341772,
479
+ "eval_loss": 1.8654963970184326,
480
+ "eval_runtime": 4.5885,
481
+ "eval_samples_per_second": 108.969,
482
+ "eval_steps_per_second": 13.73,
483
+ "step": 6535
484
+ },
485
+ {
486
+ "epoch": 6.0,
487
+ "eval_exact_match": 23.4,
488
+ "eval_f1": 31.89286435786437,
489
+ "step": 6535
490
+ },
491
+ {
492
+ "epoch": 6.06,
493
+ "learning_rate": 3e-05,
494
+ "loss": 1.0225,
495
+ "step": 6600
496
+ },
497
+ {
498
+ "epoch": 6.15,
499
+ "learning_rate": 3e-05,
500
+ "loss": 1.0118,
501
+ "step": 6700
502
+ },
503
+ {
504
+ "epoch": 6.24,
505
+ "learning_rate": 3e-05,
506
+ "loss": 1.0165,
507
+ "step": 6800
508
+ },
509
+ {
510
+ "epoch": 6.33,
511
+ "learning_rate": 3e-05,
512
+ "loss": 1.0042,
513
+ "step": 6900
514
+ },
515
+ {
516
+ "epoch": 6.43,
517
+ "learning_rate": 3e-05,
518
+ "loss": 1.0064,
519
+ "step": 7000
520
+ },
521
+ {
522
+ "epoch": 6.52,
523
+ "learning_rate": 3e-05,
524
+ "loss": 1.0245,
525
+ "step": 7100
526
+ },
527
+ {
528
+ "epoch": 6.61,
529
+ "learning_rate": 3e-05,
530
+ "loss": 1.0091,
531
+ "step": 7200
532
+ },
533
+ {
534
+ "epoch": 6.7,
535
+ "learning_rate": 3e-05,
536
+ "loss": 1.0259,
537
+ "step": 7300
538
+ },
539
+ {
540
+ "epoch": 6.79,
541
+ "learning_rate": 3e-05,
542
+ "loss": 1.0,
543
+ "step": 7400
544
+ },
545
+ {
546
+ "epoch": 6.89,
547
+ "learning_rate": 3e-05,
548
+ "loss": 0.995,
549
+ "step": 7500
550
+ },
551
+ {
552
+ "epoch": 6.98,
553
+ "learning_rate": 3e-05,
554
+ "loss": 1.0139,
555
+ "step": 7600
556
+ },
557
+ {
558
+ "epoch": 7.0,
559
+ "eval_accuracy": 0.592886075949367,
560
+ "eval_loss": 1.927990198135376,
561
+ "eval_runtime": 4.6988,
562
+ "eval_samples_per_second": 106.411,
563
+ "eval_steps_per_second": 13.408,
564
+ "step": 7624
565
+ },
566
+ {
567
+ "epoch": 7.0,
568
+ "eval_exact_match": 23.0,
569
+ "eval_f1": 32.51321067821068,
570
+ "step": 7624
571
+ },
572
+ {
573
+ "epoch": 7.07,
574
+ "learning_rate": 3e-05,
575
+ "loss": 0.9788,
576
+ "step": 7700
577
+ },
578
+ {
579
+ "epoch": 7.16,
580
+ "learning_rate": 3e-05,
581
+ "loss": 0.9597,
582
+ "step": 7800
583
+ },
584
+ {
585
+ "epoch": 7.25,
586
+ "learning_rate": 3e-05,
587
+ "loss": 0.9623,
588
+ "step": 7900
589
+ },
590
+ {
591
+ "epoch": 7.34,
592
+ "learning_rate": 3e-05,
593
+ "loss": 0.9815,
594
+ "step": 8000
595
+ },
596
+ {
597
+ "epoch": 7.44,
598
+ "learning_rate": 3e-05,
599
+ "loss": 0.9831,
600
+ "step": 8100
601
+ },
602
+ {
603
+ "epoch": 7.53,
604
+ "learning_rate": 3e-05,
605
+ "loss": 0.9762,
606
+ "step": 8200
607
+ },
608
+ {
609
+ "epoch": 7.62,
610
+ "learning_rate": 3e-05,
611
+ "loss": 0.9769,
612
+ "step": 8300
613
+ },
614
+ {
615
+ "epoch": 7.71,
616
+ "learning_rate": 3e-05,
617
+ "loss": 0.9882,
618
+ "step": 8400
619
+ },
620
+ {
621
+ "epoch": 7.8,
622
+ "learning_rate": 3e-05,
623
+ "loss": 0.9736,
624
+ "step": 8500
625
+ },
626
+ {
627
+ "epoch": 7.9,
628
+ "learning_rate": 3e-05,
629
+ "loss": 0.9764,
630
+ "step": 8600
631
+ },
632
+ {
633
+ "epoch": 7.99,
634
+ "learning_rate": 3e-05,
635
+ "loss": 0.9764,
636
+ "step": 8700
637
+ },
638
+ {
639
+ "epoch": 8.0,
640
+ "eval_accuracy": 0.5912658227848101,
641
+ "eval_loss": 1.991403579711914,
642
+ "eval_runtime": 4.6999,
643
+ "eval_samples_per_second": 106.385,
644
+ "eval_steps_per_second": 13.404,
645
+ "step": 8714
646
+ },
647
+ {
648
+ "epoch": 8.0,
649
+ "eval_exact_match": 20.8,
650
+ "eval_f1": 30.54087925962926,
651
+ "step": 8714
652
+ },
653
+ {
654
+ "epoch": 8.08,
655
+ "learning_rate": 3e-05,
656
+ "loss": 0.9509,
657
+ "step": 8800
658
+ },
659
+ {
660
+ "epoch": 8.17,
661
+ "learning_rate": 3e-05,
662
+ "loss": 0.9319,
663
+ "step": 8900
664
+ },
665
+ {
666
+ "epoch": 8.26,
667
+ "learning_rate": 3e-05,
668
+ "loss": 0.9254,
669
+ "step": 9000
670
+ },
671
+ {
672
+ "epoch": 8.35,
673
+ "learning_rate": 3e-05,
674
+ "loss": 0.9396,
675
+ "step": 9100
676
+ },
677
+ {
678
+ "epoch": 8.45,
679
+ "learning_rate": 3e-05,
680
+ "loss": 0.9439,
681
+ "step": 9200
682
+ },
683
+ {
684
+ "epoch": 8.54,
685
+ "learning_rate": 3e-05,
686
+ "loss": 0.9434,
687
+ "step": 9300
688
+ },
689
+ {
690
+ "epoch": 8.63,
691
+ "learning_rate": 3e-05,
692
+ "loss": 0.9283,
693
+ "step": 9400
694
+ },
695
+ {
696
+ "epoch": 8.72,
697
+ "learning_rate": 3e-05,
698
+ "loss": 0.9409,
699
+ "step": 9500
700
+ },
701
+ {
702
+ "epoch": 8.81,
703
+ "learning_rate": 3e-05,
704
+ "loss": 0.9474,
705
+ "step": 9600
706
+ },
707
+ {
708
+ "epoch": 8.91,
709
+ "learning_rate": 3e-05,
710
+ "loss": 0.9538,
711
+ "step": 9700
712
+ },
713
+ {
714
+ "epoch": 9.0,
715
+ "learning_rate": 3e-05,
716
+ "loss": 0.9351,
717
+ "step": 9800
718
+ },
719
+ {
720
+ "epoch": 9.0,
721
+ "eval_accuracy": 0.5909367088607594,
722
+ "eval_loss": 2.0564663410186768,
723
+ "eval_runtime": 4.8989,
724
+ "eval_samples_per_second": 102.064,
725
+ "eval_steps_per_second": 12.86,
726
+ "step": 9803
727
+ },
728
+ {
729
+ "epoch": 9.0,
730
+ "eval_exact_match": 20.2,
731
+ "eval_f1": 29.28648074148075,
732
+ "step": 9803
733
+ },
734
+ {
735
+ "epoch": 9.09,
736
+ "learning_rate": 3e-05,
737
+ "loss": 0.8896,
738
+ "step": 9900
739
+ },
740
+ {
741
+ "epoch": 9.18,
742
+ "learning_rate": 3e-05,
743
+ "loss": 0.8923,
744
+ "step": 10000
745
+ },
746
+ {
747
+ "epoch": 9.27,
748
+ "learning_rate": 3e-05,
749
+ "loss": 0.9074,
750
+ "step": 10100
751
+ },
752
+ {
753
+ "epoch": 9.36,
754
+ "learning_rate": 3e-05,
755
+ "loss": 0.905,
756
+ "step": 10200
757
+ },
758
+ {
759
+ "epoch": 9.46,
760
+ "learning_rate": 3e-05,
761
+ "loss": 0.8954,
762
+ "step": 10300
763
+ },
764
+ {
765
+ "epoch": 9.55,
766
+ "learning_rate": 3e-05,
767
+ "loss": 0.8953,
768
+ "step": 10400
769
+ },
770
+ {
771
+ "epoch": 9.64,
772
+ "learning_rate": 3e-05,
773
+ "loss": 0.8943,
774
+ "step": 10500
775
+ },
776
+ {
777
+ "epoch": 9.73,
778
+ "learning_rate": 3e-05,
779
+ "loss": 0.9173,
780
+ "step": 10600
781
+ },
782
+ {
783
+ "epoch": 9.82,
784
+ "learning_rate": 3e-05,
785
+ "loss": 0.9148,
786
+ "step": 10700
787
+ },
788
+ {
789
+ "epoch": 9.92,
790
+ "learning_rate": 3e-05,
791
+ "loss": 0.9177,
792
+ "step": 10800
793
+ },
794
+ {
795
+ "epoch": 10.0,
796
+ "eval_accuracy": 0.5891898734177216,
797
+ "eval_loss": 2.124812126159668,
798
+ "eval_runtime": 4.5237,
799
+ "eval_samples_per_second": 110.528,
800
+ "eval_steps_per_second": 13.927,
801
+ "step": 10892
802
+ },
803
+ {
804
+ "epoch": 10.0,
805
+ "eval_exact_match": 0.0,
806
+ "eval_f1": 1.653482807812982,
807
+ "step": 10892
808
+ },
809
+ {
810
+ "epoch": 10.01,
811
+ "learning_rate": 3e-05,
812
+ "loss": 0.9157,
813
+ "step": 10900
814
+ },
815
+ {
816
+ "epoch": 10.1,
817
+ "learning_rate": 3e-05,
818
+ "loss": 0.8335,
819
+ "step": 11000
820
+ },
821
+ {
822
+ "epoch": 10.19,
823
+ "learning_rate": 3e-05,
824
+ "loss": 0.8485,
825
+ "step": 11100
826
+ },
827
+ {
828
+ "epoch": 10.28,
829
+ "learning_rate": 3e-05,
830
+ "loss": 0.8712,
831
+ "step": 11200
832
+ },
833
+ {
834
+ "epoch": 10.37,
835
+ "learning_rate": 3e-05,
836
+ "loss": 0.869,
837
+ "step": 11300
838
+ },
839
+ {
840
+ "epoch": 10.47,
841
+ "learning_rate": 3e-05,
842
+ "loss": 0.8655,
843
+ "step": 11400
844
+ },
845
+ {
846
+ "epoch": 10.56,
847
+ "learning_rate": 3e-05,
848
+ "loss": 0.8868,
849
+ "step": 11500
850
+ },
851
+ {
852
+ "epoch": 10.65,
853
+ "learning_rate": 3e-05,
854
+ "loss": 0.8775,
855
+ "step": 11600
856
+ },
857
+ {
858
+ "epoch": 10.74,
859
+ "learning_rate": 3e-05,
860
+ "loss": 0.8861,
861
+ "step": 11700
862
+ },
863
+ {
864
+ "epoch": 10.83,
865
+ "learning_rate": 3e-05,
866
+ "loss": 0.892,
867
+ "step": 11800
868
+ },
869
+ {
870
+ "epoch": 10.92,
871
+ "learning_rate": 3e-05,
872
+ "loss": 0.8872,
873
+ "step": 11900
874
+ },
875
+ {
876
+ "epoch": 11.0,
877
+ "eval_accuracy": 0.5874683544303797,
878
+ "eval_loss": 2.2181763648986816,
879
+ "eval_runtime": 4.6178,
880
+ "eval_samples_per_second": 108.277,
881
+ "eval_steps_per_second": 13.643,
882
+ "step": 11981
883
+ },
884
+ {
885
+ "epoch": 11.0,
886
+ "eval_exact_match": 0.0,
887
+ "eval_f1": 0.014945652173913042,
888
+ "step": 11981
889
+ },
890
+ {
891
+ "epoch": 11.02,
892
+ "learning_rate": 3e-05,
893
+ "loss": 0.8808,
894
+ "step": 12000
895
+ },
896
+ {
897
+ "epoch": 11.11,
898
+ "learning_rate": 3e-05,
899
+ "loss": 0.8145,
900
+ "step": 12100
901
+ },
902
+ {
903
+ "epoch": 11.2,
904
+ "learning_rate": 3e-05,
905
+ "loss": 0.8301,
906
+ "step": 12200
907
+ },
908
+ {
909
+ "epoch": 11.29,
910
+ "learning_rate": 3e-05,
911
+ "loss": 0.8281,
912
+ "step": 12300
913
+ },
914
+ {
915
+ "epoch": 11.38,
916
+ "learning_rate": 3e-05,
917
+ "loss": 0.8464,
918
+ "step": 12400
919
+ },
920
+ {
921
+ "epoch": 11.48,
922
+ "learning_rate": 3e-05,
923
+ "loss": 0.8547,
924
+ "step": 12500
925
+ },
926
+ {
927
+ "epoch": 11.57,
928
+ "learning_rate": 3e-05,
929
+ "loss": 0.8424,
930
+ "step": 12600
931
+ },
932
+ {
933
+ "epoch": 11.66,
934
+ "learning_rate": 3e-05,
935
+ "loss": 0.8452,
936
+ "step": 12700
937
+ },
938
+ {
939
+ "epoch": 11.75,
940
+ "learning_rate": 3e-05,
941
+ "loss": 0.8528,
942
+ "step": 12800
943
+ },
944
+ {
945
+ "epoch": 11.84,
946
+ "learning_rate": 3e-05,
947
+ "loss": 0.862,
948
+ "step": 12900
949
+ },
950
+ {
951
+ "epoch": 11.93,
952
+ "learning_rate": 3e-05,
953
+ "loss": 0.8458,
954
+ "step": 13000
955
+ },
956
+ {
957
+ "epoch": 12.0,
958
+ "eval_accuracy": 0.5862784810126582,
959
+ "eval_loss": 2.286332130432129,
960
+ "eval_runtime": 4.5481,
961
+ "eval_samples_per_second": 109.935,
962
+ "eval_steps_per_second": 13.852,
963
+ "step": 13071
964
+ },
965
+ {
966
+ "epoch": 12.0,
967
+ "eval_exact_match": 0.0,
968
+ "eval_f1": 0.05640597466324206,
969
+ "step": 13071
970
+ },
971
+ {
972
+ "epoch": 12.03,
973
+ "learning_rate": 3e-05,
974
+ "loss": 0.8407,
975
+ "step": 13100
976
+ },
977
+ {
978
+ "epoch": 12.12,
979
+ "learning_rate": 3e-05,
980
+ "loss": 0.8103,
981
+ "step": 13200
982
+ },
983
+ {
984
+ "epoch": 12.21,
985
+ "learning_rate": 3e-05,
986
+ "loss": 0.8044,
987
+ "step": 13300
988
+ },
989
+ {
990
+ "epoch": 12.3,
991
+ "learning_rate": 3e-05,
992
+ "loss": 0.819,
993
+ "step": 13400
994
+ },
995
+ {
996
+ "epoch": 12.39,
997
+ "learning_rate": 3e-05,
998
+ "loss": 0.7983,
999
+ "step": 13500
1000
+ },
1001
+ {
1002
+ "epoch": 12.49,
1003
+ "learning_rate": 3e-05,
1004
+ "loss": 0.8057,
1005
+ "step": 13600
1006
+ },
1007
+ {
1008
+ "epoch": 12.58,
1009
+ "learning_rate": 3e-05,
1010
+ "loss": 0.8133,
1011
+ "step": 13700
1012
+ },
1013
+ {
1014
+ "epoch": 12.67,
1015
+ "learning_rate": 3e-05,
1016
+ "loss": 0.8267,
1017
+ "step": 13800
1018
+ },
1019
+ {
1020
+ "epoch": 12.76,
1021
+ "learning_rate": 3e-05,
1022
+ "loss": 0.8082,
1023
+ "step": 13900
1024
+ },
1025
+ {
1026
+ "epoch": 12.85,
1027
+ "learning_rate": 3e-05,
1028
+ "loss": 0.8316,
1029
+ "step": 14000
1030
+ },
1031
+ {
1032
+ "epoch": 12.94,
1033
+ "learning_rate": 3e-05,
1034
+ "loss": 0.8148,
1035
+ "step": 14100
1036
+ },
1037
+ {
1038
+ "epoch": 13.0,
1039
+ "eval_accuracy": 0.5842278481012658,
1040
+ "eval_loss": 2.352457284927368,
1041
+ "eval_runtime": 4.464,
1042
+ "eval_samples_per_second": 112.008,
1043
+ "eval_steps_per_second": 14.113,
1044
+ "step": 14160
1045
+ },
1046
+ {
1047
+ "epoch": 13.0,
1048
+ "eval_exact_match": 0.0,
1049
+ "eval_f1": 0.08618200930627685,
1050
+ "step": 14160
1051
+ },
1052
+ {
1053
+ "epoch": 13.04,
1054
+ "learning_rate": 3e-05,
1055
+ "loss": 0.798,
1056
+ "step": 14200
1057
+ },
1058
+ {
1059
+ "epoch": 13.13,
1060
+ "learning_rate": 3e-05,
1061
+ "loss": 0.7788,
1062
+ "step": 14300
1063
+ },
1064
+ {
1065
+ "epoch": 13.22,
1066
+ "learning_rate": 3e-05,
1067
+ "loss": 0.7727,
1068
+ "step": 14400
1069
+ },
1070
+ {
1071
+ "epoch": 13.31,
1072
+ "learning_rate": 3e-05,
1073
+ "loss": 0.7831,
1074
+ "step": 14500
1075
+ },
1076
+ {
1077
+ "epoch": 13.4,
1078
+ "learning_rate": 3e-05,
1079
+ "loss": 0.7901,
1080
+ "step": 14600
1081
+ },
1082
+ {
1083
+ "epoch": 13.5,
1084
+ "learning_rate": 3e-05,
1085
+ "loss": 0.7993,
1086
+ "step": 14700
1087
+ },
1088
+ {
1089
+ "epoch": 13.59,
1090
+ "learning_rate": 3e-05,
1091
+ "loss": 0.7864,
1092
+ "step": 14800
1093
+ },
1094
+ {
1095
+ "epoch": 13.68,
1096
+ "learning_rate": 3e-05,
1097
+ "loss": 0.7813,
1098
+ "step": 14900
1099
+ },
1100
+ {
1101
+ "epoch": 13.77,
1102
+ "learning_rate": 3e-05,
1103
+ "loss": 0.8024,
1104
+ "step": 15000
1105
+ },
1106
+ {
1107
+ "epoch": 13.86,
1108
+ "learning_rate": 3e-05,
1109
+ "loss": 0.7945,
1110
+ "step": 15100
1111
+ },
1112
+ {
1113
+ "epoch": 13.95,
1114
+ "learning_rate": 3e-05,
1115
+ "loss": 0.7955,
1116
+ "step": 15200
1117
+ },
1118
+ {
1119
+ "epoch": 14.0,
1120
+ "eval_accuracy": 0.5834177215189873,
1121
+ "eval_loss": 2.444202184677124,
1122
+ "eval_runtime": 4.5506,
1123
+ "eval_samples_per_second": 109.877,
1124
+ "eval_steps_per_second": 13.844,
1125
+ "step": 15249
1126
+ },
1127
+ {
1128
+ "epoch": 14.0,
1129
+ "eval_exact_match": 0.0,
1130
+ "eval_f1": 0.04849394305657365,
1131
+ "step": 15249
1132
+ },
1133
+ {
1134
+ "epoch": 14.05,
1135
+ "learning_rate": 3e-05,
1136
+ "loss": 0.7808,
1137
+ "step": 15300
1138
+ },
1139
+ {
1140
+ "epoch": 14.14,
1141
+ "learning_rate": 3e-05,
1142
+ "loss": 0.752,
1143
+ "step": 15400
1144
+ },
1145
+ {
1146
+ "epoch": 14.23,
1147
+ "learning_rate": 3e-05,
1148
+ "loss": 0.7533,
1149
+ "step": 15500
1150
+ },
1151
+ {
1152
+ "epoch": 14.32,
1153
+ "learning_rate": 3e-05,
1154
+ "loss": 0.7458,
1155
+ "step": 15600
1156
+ },
1157
+ {
1158
+ "epoch": 14.41,
1159
+ "learning_rate": 3e-05,
1160
+ "loss": 0.7527,
1161
+ "step": 15700
1162
+ },
1163
+ {
1164
+ "epoch": 14.51,
1165
+ "learning_rate": 3e-05,
1166
+ "loss": 0.7625,
1167
+ "step": 15800
1168
+ },
1169
+ {
1170
+ "epoch": 14.6,
1171
+ "learning_rate": 3e-05,
1172
+ "loss": 0.7424,
1173
+ "step": 15900
1174
+ },
1175
+ {
1176
+ "epoch": 14.69,
1177
+ "learning_rate": 3e-05,
1178
+ "loss": 0.7618,
1179
+ "step": 16000
1180
+ },
1181
+ {
1182
+ "epoch": 14.78,
1183
+ "learning_rate": 3e-05,
1184
+ "loss": 0.7589,
1185
+ "step": 16100
1186
+ },
1187
+ {
1188
+ "epoch": 14.87,
1189
+ "learning_rate": 3e-05,
1190
+ "loss": 0.7657,
1191
+ "step": 16200
1192
+ },
1193
+ {
1194
+ "epoch": 14.96,
1195
+ "learning_rate": 3e-05,
1196
+ "loss": 0.7765,
1197
+ "step": 16300
1198
+ },
1199
+ {
1200
+ "epoch": 15.0,
1201
+ "eval_accuracy": 0.583620253164557,
1202
+ "eval_loss": 2.4962217807769775,
1203
+ "eval_runtime": 4.7551,
1204
+ "eval_samples_per_second": 105.15,
1205
+ "eval_steps_per_second": 13.249,
1206
+ "step": 16338
1207
+ },
1208
+ {
1209
+ "epoch": 15.0,
1210
+ "eval_exact_match": 0.0,
1211
+ "eval_f1": 0.013684210526315792,
1212
+ "step": 16338
1213
+ },
1214
+ {
1215
+ "epoch": 15.06,
1216
+ "learning_rate": 3e-05,
1217
+ "loss": 0.7377,
1218
+ "step": 16400
1219
+ },
1220
+ {
1221
+ "epoch": 15.15,
1222
+ "learning_rate": 3e-05,
1223
+ "loss": 0.7204,
1224
+ "step": 16500
1225
+ },
1226
+ {
1227
+ "epoch": 15.24,
1228
+ "learning_rate": 3e-05,
1229
+ "loss": 0.7176,
1230
+ "step": 16600
1231
+ },
1232
+ {
1233
+ "epoch": 15.33,
1234
+ "learning_rate": 3e-05,
1235
+ "loss": 0.7164,
1236
+ "step": 16700
1237
+ },
1238
+ {
1239
+ "epoch": 15.42,
1240
+ "learning_rate": 3e-05,
1241
+ "loss": 0.7266,
1242
+ "step": 16800
1243
+ },
1244
+ {
1245
+ "epoch": 15.52,
1246
+ "learning_rate": 3e-05,
1247
+ "loss": 0.7393,
1248
+ "step": 16900
1249
+ },
1250
+ {
1251
+ "epoch": 15.61,
1252
+ "learning_rate": 3e-05,
1253
+ "loss": 0.727,
1254
+ "step": 17000
1255
+ },
1256
+ {
1257
+ "epoch": 15.7,
1258
+ "learning_rate": 3e-05,
1259
+ "loss": 0.7364,
1260
+ "step": 17100
1261
+ },
1262
+ {
1263
+ "epoch": 15.79,
1264
+ "learning_rate": 3e-05,
1265
+ "loss": 0.7337,
1266
+ "step": 17200
1267
+ },
1268
+ {
1269
+ "epoch": 15.88,
1270
+ "learning_rate": 3e-05,
1271
+ "loss": 0.7458,
1272
+ "step": 17300
1273
+ },
1274
+ {
1275
+ "epoch": 15.97,
1276
+ "learning_rate": 3e-05,
1277
+ "loss": 0.7412,
1278
+ "step": 17400
1279
+ },
1280
+ {
1281
+ "epoch": 16.0,
1282
+ "eval_accuracy": 0.5817974683544304,
1283
+ "eval_loss": 2.574023962020874,
1284
+ "eval_runtime": 4.5479,
1285
+ "eval_samples_per_second": 109.941,
1286
+ "eval_steps_per_second": 13.853,
1287
+ "step": 17428
1288
+ },
1289
+ {
1290
+ "epoch": 16.0,
1291
+ "eval_exact_match": 0.0,
1292
+ "eval_f1": 0.006153846153846153,
1293
+ "step": 17428
1294
+ },
1295
+ {
1296
+ "epoch": 16.07,
1297
+ "learning_rate": 3e-05,
1298
+ "loss": 0.6984,
1299
+ "step": 17500
1300
+ },
1301
+ {
1302
+ "epoch": 16.16,
1303
+ "learning_rate": 3e-05,
1304
+ "loss": 0.6995,
1305
+ "step": 17600
1306
+ },
1307
+ {
1308
+ "epoch": 16.25,
1309
+ "learning_rate": 3e-05,
1310
+ "loss": 0.6936,
1311
+ "step": 17700
1312
+ },
1313
+ {
1314
+ "epoch": 16.34,
1315
+ "learning_rate": 3e-05,
1316
+ "loss": 0.7046,
1317
+ "step": 17800
1318
+ },
1319
+ {
1320
+ "epoch": 16.43,
1321
+ "learning_rate": 3e-05,
1322
+ "loss": 0.7088,
1323
+ "step": 17900
1324
+ },
1325
+ {
1326
+ "epoch": 16.53,
1327
+ "learning_rate": 3e-05,
1328
+ "loss": 0.7065,
1329
+ "step": 18000
1330
+ },
1331
+ {
1332
+ "epoch": 16.62,
1333
+ "learning_rate": 3e-05,
1334
+ "loss": 0.7009,
1335
+ "step": 18100
1336
+ },
1337
+ {
1338
+ "epoch": 16.71,
1339
+ "learning_rate": 3e-05,
1340
+ "loss": 0.7106,
1341
+ "step": 18200
1342
+ },
1343
+ {
1344
+ "epoch": 16.8,
1345
+ "learning_rate": 3e-05,
1346
+ "loss": 0.7131,
1347
+ "step": 18300
1348
+ },
1349
+ {
1350
+ "epoch": 16.89,
1351
+ "learning_rate": 3e-05,
1352
+ "loss": 0.6996,
1353
+ "step": 18400
1354
+ },
1355
+ {
1356
+ "epoch": 16.98,
1357
+ "learning_rate": 3e-05,
1358
+ "loss": 0.7257,
1359
+ "step": 18500
1360
+ },
1361
+ {
1362
+ "epoch": 17.0,
1363
+ "eval_accuracy": 0.5816708860759494,
1364
+ "eval_loss": 2.653719425201416,
1365
+ "eval_runtime": 5.1182,
1366
+ "eval_samples_per_second": 97.691,
1367
+ "eval_steps_per_second": 12.309,
1368
+ "step": 18517
1369
+ },
1370
+ {
1371
+ "epoch": 17.0,
1372
+ "eval_exact_match": 0.0,
1373
+ "eval_f1": 0.012886859462201928,
1374
+ "step": 18517
1375
+ },
1376
+ {
1377
+ "epoch": 17.08,
1378
+ "learning_rate": 3e-05,
1379
+ "loss": 0.6612,
1380
+ "step": 18600
1381
+ },
1382
+ {
1383
+ "epoch": 17.17,
1384
+ "learning_rate": 3e-05,
1385
+ "loss": 0.6697,
1386
+ "step": 18700
1387
+ },
1388
+ {
1389
+ "epoch": 17.26,
1390
+ "learning_rate": 3e-05,
1391
+ "loss": 0.674,
1392
+ "step": 18800
1393
+ },
1394
+ {
1395
+ "epoch": 17.35,
1396
+ "learning_rate": 3e-05,
1397
+ "loss": 0.6737,
1398
+ "step": 18900
1399
+ },
1400
+ {
1401
+ "epoch": 17.44,
1402
+ "learning_rate": 3e-05,
1403
+ "loss": 0.6946,
1404
+ "step": 19000
1405
+ },
1406
+ {
1407
+ "epoch": 17.54,
1408
+ "learning_rate": 3e-05,
1409
+ "loss": 0.684,
1410
+ "step": 19100
1411
+ },
1412
+ {
1413
+ "epoch": 17.63,
1414
+ "learning_rate": 3e-05,
1415
+ "loss": 0.6888,
1416
+ "step": 19200
1417
+ },
1418
+ {
1419
+ "epoch": 17.72,
1420
+ "learning_rate": 3e-05,
1421
+ "loss": 0.6914,
1422
+ "step": 19300
1423
+ },
1424
+ {
1425
+ "epoch": 17.81,
1426
+ "learning_rate": 3e-05,
1427
+ "loss": 0.697,
1428
+ "step": 19400
1429
+ },
1430
+ {
1431
+ "epoch": 17.9,
1432
+ "learning_rate": 3e-05,
1433
+ "loss": 0.7006,
1434
+ "step": 19500
1435
+ },
1436
+ {
1437
+ "epoch": 17.99,
1438
+ "learning_rate": 3e-05,
1439
+ "loss": 0.7085,
1440
+ "step": 19600
1441
+ },
1442
+ {
1443
+ "epoch": 18.0,
1444
+ "eval_accuracy": 0.5813417721518988,
1445
+ "eval_loss": 2.6944408416748047,
1446
+ "eval_runtime": 4.883,
1447
+ "eval_samples_per_second": 102.397,
1448
+ "eval_steps_per_second": 12.902,
1449
+ "step": 19606
1450
+ },
1451
+ {
1452
+ "epoch": 18.0,
1453
+ "eval_exact_match": 0.0,
1454
+ "eval_f1": 0.0,
1455
+ "step": 19606
1456
+ },
1457
+ {
1458
+ "epoch": 18.09,
1459
+ "learning_rate": 3e-05,
1460
+ "loss": 0.6696,
1461
+ "step": 19700
1462
+ },
1463
+ {
1464
+ "epoch": 18.18,
1465
+ "learning_rate": 3e-05,
1466
+ "loss": 0.656,
1467
+ "step": 19800
1468
+ },
1469
+ {
1470
+ "epoch": 18.27,
1471
+ "learning_rate": 3e-05,
1472
+ "loss": 0.66,
1473
+ "step": 19900
1474
+ },
1475
+ {
1476
+ "epoch": 18.36,
1477
+ "learning_rate": 3e-05,
1478
+ "loss": 0.6609,
1479
+ "step": 20000
1480
+ },
1481
+ {
1482
+ "epoch": 18.45,
1483
+ "learning_rate": 3e-05,
1484
+ "loss": 0.6616,
1485
+ "step": 20100
1486
+ },
1487
+ {
1488
+ "epoch": 18.54,
1489
+ "learning_rate": 3e-05,
1490
+ "loss": 0.6743,
1491
+ "step": 20200
1492
+ },
1493
+ {
1494
+ "epoch": 18.64,
1495
+ "learning_rate": 3e-05,
1496
+ "loss": 0.6721,
1497
+ "step": 20300
1498
+ },
1499
+ {
1500
+ "epoch": 18.73,
1501
+ "learning_rate": 3e-05,
1502
+ "loss": 0.6715,
1503
+ "step": 20400
1504
+ },
1505
+ {
1506
+ "epoch": 18.82,
1507
+ "learning_rate": 3e-05,
1508
+ "loss": 0.678,
1509
+ "step": 20500
1510
+ },
1511
+ {
1512
+ "epoch": 18.91,
1513
+ "learning_rate": 3e-05,
1514
+ "loss": 0.6756,
1515
+ "step": 20600
1516
+ },
1517
+ {
1518
+ "epoch": 19.0,
1519
+ "eval_accuracy": 0.5794683544303797,
1520
+ "eval_loss": 2.761476755142212,
1521
+ "eval_runtime": 4.6988,
1522
+ "eval_samples_per_second": 106.411,
1523
+ "eval_steps_per_second": 13.408,
1524
+ "step": 20695
1525
+ },
1526
+ {
1527
+ "epoch": 19.0,
1528
+ "eval_exact_match": 0.0,
1529
+ "eval_f1": 0.0,
1530
+ "step": 20695
1531
+ },
1532
+ {
1533
+ "epoch": 19.0,
1534
+ "learning_rate": 3e-05,
1535
+ "loss": 0.6761,
1536
+ "step": 20700
1537
+ },
1538
+ {
1539
+ "epoch": 19.1,
1540
+ "learning_rate": 3e-05,
1541
+ "loss": 0.6354,
1542
+ "step": 20800
1543
+ },
1544
+ {
1545
+ "epoch": 19.19,
1546
+ "learning_rate": 3e-05,
1547
+ "loss": 0.6472,
1548
+ "step": 20900
1549
+ },
1550
+ {
1551
+ "epoch": 19.28,
1552
+ "learning_rate": 3e-05,
1553
+ "loss": 0.638,
1554
+ "step": 21000
1555
+ },
1556
+ {
1557
+ "epoch": 19.37,
1558
+ "learning_rate": 3e-05,
1559
+ "loss": 0.6251,
1560
+ "step": 21100
1561
+ },
1562
+ {
1563
+ "epoch": 19.46,
1564
+ "learning_rate": 3e-05,
1565
+ "loss": 0.6472,
1566
+ "step": 21200
1567
+ },
1568
+ {
1569
+ "epoch": 19.55,
1570
+ "learning_rate": 3e-05,
1571
+ "loss": 0.6459,
1572
+ "step": 21300
1573
+ },
1574
+ {
1575
+ "epoch": 19.65,
1576
+ "learning_rate": 3e-05,
1577
+ "loss": 0.6473,
1578
+ "step": 21400
1579
+ },
1580
+ {
1581
+ "epoch": 19.74,
1582
+ "learning_rate": 3e-05,
1583
+ "loss": 0.6492,
1584
+ "step": 21500
1585
+ },
1586
+ {
1587
+ "epoch": 19.83,
1588
+ "learning_rate": 3e-05,
1589
+ "loss": 0.6625,
1590
+ "step": 21600
1591
+ },
1592
+ {
1593
+ "epoch": 19.92,
1594
+ "learning_rate": 3e-05,
1595
+ "loss": 0.6493,
1596
+ "step": 21700
1597
+ },
1598
+ {
1599
+ "epoch": 20.0,
1600
+ "eval_accuracy": 0.5800506329113924,
1601
+ "eval_loss": 2.8422911167144775,
1602
+ "eval_runtime": 4.4906,
1603
+ "eval_samples_per_second": 111.343,
1604
+ "eval_steps_per_second": 14.029,
1605
+ "step": 21785
1606
+ },
1607
+ {
1608
+ "epoch": 20.0,
1609
+ "eval_exact_match": 0.0,
1610
+ "eval_f1": 0.0,
1611
+ "step": 21785
1612
+ },
1613
+ {
1614
+ "epoch": 20.01,
1615
+ "learning_rate": 3e-05,
1616
+ "loss": 0.644,
1617
+ "step": 21800
1618
+ },
1619
+ {
1620
+ "epoch": 20.11,
1621
+ "learning_rate": 3e-05,
1622
+ "loss": 0.6127,
1623
+ "step": 21900
1624
+ },
1625
+ {
1626
+ "epoch": 20.2,
1627
+ "learning_rate": 3e-05,
1628
+ "loss": 0.6111,
1629
+ "step": 22000
1630
+ },
1631
+ {
1632
+ "epoch": 20.29,
1633
+ "learning_rate": 3e-05,
1634
+ "loss": 0.6231,
1635
+ "step": 22100
1636
+ },
1637
+ {
1638
+ "epoch": 20.38,
1639
+ "learning_rate": 3e-05,
1640
+ "loss": 0.6229,
1641
+ "step": 22200
1642
+ },
1643
+ {
1644
+ "epoch": 20.47,
1645
+ "learning_rate": 3e-05,
1646
+ "loss": 0.6215,
1647
+ "step": 22300
1648
+ },
1649
+ {
1650
+ "epoch": 20.56,
1651
+ "learning_rate": 3e-05,
1652
+ "loss": 0.644,
1653
+ "step": 22400
1654
+ },
1655
+ {
1656
+ "epoch": 20.66,
1657
+ "learning_rate": 3e-05,
1658
+ "loss": 0.6501,
1659
+ "step": 22500
1660
+ },
1661
+ {
1662
+ "epoch": 20.75,
1663
+ "learning_rate": 3e-05,
1664
+ "loss": 0.6425,
1665
+ "step": 22600
1666
+ },
1667
+ {
1668
+ "epoch": 20.84,
1669
+ "learning_rate": 3e-05,
1670
+ "loss": 0.6421,
1671
+ "step": 22700
1672
+ },
1673
+ {
1674
+ "epoch": 20.93,
1675
+ "learning_rate": 3e-05,
1676
+ "loss": 0.6545,
1677
+ "step": 22800
1678
+ },
1679
+ {
1680
+ "epoch": 21.0,
1681
+ "eval_accuracy": 0.5791645569620253,
1682
+ "eval_loss": 2.8643393516540527,
1683
+ "eval_runtime": 5.1135,
1684
+ "eval_samples_per_second": 97.781,
1685
+ "eval_steps_per_second": 12.32,
1686
+ "step": 22874
1687
+ },
1688
+ {
1689
+ "epoch": 21.0,
1690
+ "eval_exact_match": 0.0,
1691
+ "eval_f1": 0.0,
1692
+ "step": 22874
1693
+ },
1694
+ {
1695
+ "epoch": 21.02,
1696
+ "learning_rate": 3e-05,
1697
+ "loss": 0.6434,
1698
+ "step": 22900
1699
+ },
1700
+ {
1701
+ "epoch": 21.12,
1702
+ "learning_rate": 3e-05,
1703
+ "loss": 0.6251,
1704
+ "step": 23000
1705
+ },
1706
+ {
1707
+ "epoch": 21.21,
1708
+ "learning_rate": 3e-05,
1709
+ "loss": 0.6239,
1710
+ "step": 23100
1711
+ },
1712
+ {
1713
+ "epoch": 21.3,
1714
+ "learning_rate": 3e-05,
1715
+ "loss": 0.6287,
1716
+ "step": 23200
1717
+ },
1718
+ {
1719
+ "epoch": 21.39,
1720
+ "learning_rate": 3e-05,
1721
+ "loss": 0.6219,
1722
+ "step": 23300
1723
+ },
1724
+ {
1725
+ "epoch": 21.48,
1726
+ "learning_rate": 3e-05,
1727
+ "loss": 0.6293,
1728
+ "step": 23400
1729
+ },
1730
+ {
1731
+ "epoch": 21.57,
1732
+ "learning_rate": 3e-05,
1733
+ "loss": 0.6407,
1734
+ "step": 23500
1735
+ },
1736
+ {
1737
+ "epoch": 21.67,
1738
+ "learning_rate": 3e-05,
1739
+ "loss": 0.6551,
1740
+ "step": 23600
1741
+ },
1742
+ {
1743
+ "epoch": 21.76,
1744
+ "learning_rate": 3e-05,
1745
+ "loss": 0.6661,
1746
+ "step": 23700
1747
+ },
1748
+ {
1749
+ "epoch": 21.85,
1750
+ "learning_rate": 3e-05,
1751
+ "loss": 0.6747,
1752
+ "step": 23800
1753
+ },
1754
+ {
1755
+ "epoch": 21.94,
1756
+ "learning_rate": 3e-05,
1757
+ "loss": 0.6726,
1758
+ "step": 23900
1759
+ },
1760
+ {
1761
+ "epoch": 22.0,
1762
+ "eval_accuracy": 0.5789620253164557,
1763
+ "eval_loss": 2.8472466468811035,
1764
+ "eval_runtime": 4.4563,
1765
+ "eval_samples_per_second": 112.201,
1766
+ "eval_steps_per_second": 14.137,
1767
+ "step": 23963
1768
+ },
1769
+ {
1770
+ "epoch": 22.0,
1771
+ "eval_exact_match": 0.0,
1772
+ "eval_f1": 0.0,
1773
+ "step": 23963
1774
+ },
1775
+ {
1776
+ "epoch": 22.03,
1777
+ "learning_rate": 3e-05,
1778
+ "loss": 0.666,
1779
+ "step": 24000
1780
+ },
1781
+ {
1782
+ "epoch": 22.13,
1783
+ "learning_rate": 3e-05,
1784
+ "loss": 0.6542,
1785
+ "step": 24100
1786
+ },
1787
+ {
1788
+ "epoch": 22.22,
1789
+ "learning_rate": 3e-05,
1790
+ "loss": 0.6506,
1791
+ "step": 24200
1792
+ },
1793
+ {
1794
+ "epoch": 22.31,
1795
+ "learning_rate": 3e-05,
1796
+ "loss": 0.644,
1797
+ "step": 24300
1798
+ },
1799
+ {
1800
+ "epoch": 22.4,
1801
+ "learning_rate": 3e-05,
1802
+ "loss": 0.6493,
1803
+ "step": 24400
1804
+ },
1805
+ {
1806
+ "epoch": 22.49,
1807
+ "learning_rate": 3e-05,
1808
+ "loss": 0.6674,
1809
+ "step": 24500
1810
+ },
1811
+ {
1812
+ "epoch": 22.58,
1813
+ "learning_rate": 3e-05,
1814
+ "loss": 0.6761,
1815
+ "step": 24600
1816
+ },
1817
+ {
1818
+ "epoch": 22.68,
1819
+ "learning_rate": 3e-05,
1820
+ "loss": 0.6841,
1821
+ "step": 24700
1822
+ },
1823
+ {
1824
+ "epoch": 22.77,
1825
+ "learning_rate": 3e-05,
1826
+ "loss": 0.6891,
1827
+ "step": 24800
1828
+ },
1829
+ {
1830
+ "epoch": 22.86,
1831
+ "learning_rate": 3e-05,
1832
+ "loss": 0.6518,
1833
+ "step": 24900
1834
+ },
1835
+ {
1836
+ "epoch": 22.95,
1837
+ "learning_rate": 3e-05,
1838
+ "loss": 0.6802,
1839
+ "step": 25000
1840
+ },
1841
+ {
1842
+ "epoch": 23.0,
1843
+ "eval_accuracy": 0.5804556962025317,
1844
+ "eval_loss": 2.8001668453216553,
1845
+ "eval_runtime": 4.7002,
1846
+ "eval_samples_per_second": 106.378,
1847
+ "eval_steps_per_second": 13.404,
1848
+ "step": 25052
1849
+ },
1850
+ {
1851
+ "epoch": 23.0,
1852
+ "eval_exact_match": 0.0,
1853
+ "eval_f1": 0.0,
1854
+ "step": 25052
1855
+ }
1856
+ ],
1857
+ "logging_steps": 100,
1858
+ "max_steps": 54450,
1859
+ "num_train_epochs": 50,
1860
+ "save_steps": 500,
1861
+ "total_flos": 4.715808218624819e+18,
1862
+ "trial_name": null,
1863
+ "trial_params": null
1864
+ }
checkpoint-25052/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:990ba0ec6800794569e1f142d0a98c1c27c344ee04644eb9619680965b92e617
3
+ size 4728