tyzhu commited on
Commit
05a1a55
1 Parent(s): e5e3d8a

Training in progress, epoch 40, checkpoint

Browse files
checkpoint-21160/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-21160/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-21160/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91f2494d3bbcef0793c81cee1817a4a07a3a8529828de25aa69217b79fd84bfb
3
+ size 143269386
checkpoint-21160/added_tokens.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "</s>": 2,
3
+ "<s>": 1,
4
+ "<unk>": 0
5
+ }
checkpoint-21160/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05b28191cd5d79d632efdda41ee3f11e4d0c8c7bd9793e4643c96a239362b8b8
3
+ size 286585234
checkpoint-21160/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d07d7b34a7831544ddcde211c84acb64aef08130fcae223c16d97db479ee4c5e
3
+ size 15024
checkpoint-21160/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ffed6debebb1ad18fa3399e3096a298a9e09f3eb9fdb35c2398d8a7df63b7737
3
+ size 15024
checkpoint-21160/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f24d64bbf427c3244c88d8e9be36dc9e5ddb55e82743ad71613d1ac490ca8d51
3
+ size 15024
checkpoint-21160/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9203a646de7d64bc0223c1e3b63595ef9da24884c6b4cc9db23d627857264bc
3
+ size 15024
checkpoint-21160/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59b5e41c3b8d28c10856390e4b0333ef4d391c26a9cdc2e12d8f9c297442e5c1
3
+ size 1064
checkpoint-21160/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-21160/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-21160/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-21160/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-21160/trainer_state.json ADDED
@@ -0,0 +1,1885 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 40.0,
5
+ "eval_steps": 500,
6
+ "global_step": 21160,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.19,
13
+ "learning_rate": 3e-05,
14
+ "loss": 1.7194,
15
+ "step": 100
16
+ },
17
+ {
18
+ "epoch": 0.38,
19
+ "learning_rate": 3e-05,
20
+ "loss": 1.4369,
21
+ "step": 200
22
+ },
23
+ {
24
+ "epoch": 0.57,
25
+ "learning_rate": 3e-05,
26
+ "loss": 1.4257,
27
+ "step": 300
28
+ },
29
+ {
30
+ "epoch": 0.76,
31
+ "learning_rate": 3e-05,
32
+ "loss": 1.3947,
33
+ "step": 400
34
+ },
35
+ {
36
+ "epoch": 0.95,
37
+ "learning_rate": 3e-05,
38
+ "loss": 1.3948,
39
+ "step": 500
40
+ },
41
+ {
42
+ "epoch": 1.0,
43
+ "eval_accuracy": 0.6132307692307692,
44
+ "eval_loss": 1.3087366819381714,
45
+ "eval_runtime": 4.3553,
46
+ "eval_samples_per_second": 114.803,
47
+ "eval_steps_per_second": 14.465,
48
+ "step": 529
49
+ },
50
+ {
51
+ "epoch": 1.0,
52
+ "eval_exact_match": 24.4,
53
+ "eval_f1": 37.614420394420414,
54
+ "step": 529
55
+ },
56
+ {
57
+ "epoch": 1.13,
58
+ "learning_rate": 3e-05,
59
+ "loss": 1.3927,
60
+ "step": 600
61
+ },
62
+ {
63
+ "epoch": 1.32,
64
+ "learning_rate": 3e-05,
65
+ "loss": 1.3644,
66
+ "step": 700
67
+ },
68
+ {
69
+ "epoch": 1.51,
70
+ "learning_rate": 3e-05,
71
+ "loss": 1.3791,
72
+ "step": 800
73
+ },
74
+ {
75
+ "epoch": 1.7,
76
+ "learning_rate": 3e-05,
77
+ "loss": 1.3627,
78
+ "step": 900
79
+ },
80
+ {
81
+ "epoch": 1.89,
82
+ "learning_rate": 3e-05,
83
+ "loss": 1.3789,
84
+ "step": 1000
85
+ },
86
+ {
87
+ "epoch": 2.0,
88
+ "eval_accuracy": 0.6146153846153846,
89
+ "eval_loss": 1.289727807044983,
90
+ "eval_runtime": 4.4281,
91
+ "eval_samples_per_second": 112.915,
92
+ "eval_steps_per_second": 14.227,
93
+ "step": 1058
94
+ },
95
+ {
96
+ "epoch": 2.0,
97
+ "eval_exact_match": 23.2,
98
+ "eval_f1": 35.56775335775336,
99
+ "step": 1058
100
+ },
101
+ {
102
+ "epoch": 2.08,
103
+ "learning_rate": 3e-05,
104
+ "loss": 1.3605,
105
+ "step": 1100
106
+ },
107
+ {
108
+ "epoch": 2.27,
109
+ "learning_rate": 3e-05,
110
+ "loss": 1.3306,
111
+ "step": 1200
112
+ },
113
+ {
114
+ "epoch": 2.46,
115
+ "learning_rate": 3e-05,
116
+ "loss": 1.3318,
117
+ "step": 1300
118
+ },
119
+ {
120
+ "epoch": 2.65,
121
+ "learning_rate": 3e-05,
122
+ "loss": 1.3491,
123
+ "step": 1400
124
+ },
125
+ {
126
+ "epoch": 2.84,
127
+ "learning_rate": 3e-05,
128
+ "loss": 1.3259,
129
+ "step": 1500
130
+ },
131
+ {
132
+ "epoch": 3.0,
133
+ "eval_accuracy": 0.6178974358974358,
134
+ "eval_loss": 1.2849316596984863,
135
+ "eval_runtime": 4.875,
136
+ "eval_samples_per_second": 102.565,
137
+ "eval_steps_per_second": 12.923,
138
+ "step": 1587
139
+ },
140
+ {
141
+ "epoch": 3.0,
142
+ "eval_exact_match": 25.2,
143
+ "eval_f1": 37.32698412698414,
144
+ "step": 1587
145
+ },
146
+ {
147
+ "epoch": 3.02,
148
+ "learning_rate": 3e-05,
149
+ "loss": 1.3255,
150
+ "step": 1600
151
+ },
152
+ {
153
+ "epoch": 3.21,
154
+ "learning_rate": 3e-05,
155
+ "loss": 1.28,
156
+ "step": 1700
157
+ },
158
+ {
159
+ "epoch": 3.4,
160
+ "learning_rate": 3e-05,
161
+ "loss": 1.2995,
162
+ "step": 1800
163
+ },
164
+ {
165
+ "epoch": 3.59,
166
+ "learning_rate": 3e-05,
167
+ "loss": 1.2991,
168
+ "step": 1900
169
+ },
170
+ {
171
+ "epoch": 3.78,
172
+ "learning_rate": 3e-05,
173
+ "loss": 1.3024,
174
+ "step": 2000
175
+ },
176
+ {
177
+ "epoch": 3.97,
178
+ "learning_rate": 3e-05,
179
+ "loss": 1.2853,
180
+ "step": 2100
181
+ },
182
+ {
183
+ "epoch": 4.0,
184
+ "eval_accuracy": 0.615948717948718,
185
+ "eval_loss": 1.316924810409546,
186
+ "eval_runtime": 5.3385,
187
+ "eval_samples_per_second": 93.659,
188
+ "eval_steps_per_second": 11.801,
189
+ "step": 2116
190
+ },
191
+ {
192
+ "epoch": 4.0,
193
+ "eval_exact_match": 23.8,
194
+ "eval_f1": 36.284920634920645,
195
+ "step": 2116
196
+ },
197
+ {
198
+ "epoch": 4.16,
199
+ "learning_rate": 3e-05,
200
+ "loss": 1.2575,
201
+ "step": 2200
202
+ },
203
+ {
204
+ "epoch": 4.35,
205
+ "learning_rate": 3e-05,
206
+ "loss": 1.2261,
207
+ "step": 2300
208
+ },
209
+ {
210
+ "epoch": 4.54,
211
+ "learning_rate": 3e-05,
212
+ "loss": 1.2536,
213
+ "step": 2400
214
+ },
215
+ {
216
+ "epoch": 4.73,
217
+ "learning_rate": 3e-05,
218
+ "loss": 1.2376,
219
+ "step": 2500
220
+ },
221
+ {
222
+ "epoch": 4.91,
223
+ "learning_rate": 3e-05,
224
+ "loss": 1.2556,
225
+ "step": 2600
226
+ },
227
+ {
228
+ "epoch": 5.0,
229
+ "eval_accuracy": 0.6131794871794872,
230
+ "eval_loss": 1.3531945943832397,
231
+ "eval_runtime": 4.4382,
232
+ "eval_samples_per_second": 112.659,
233
+ "eval_steps_per_second": 14.195,
234
+ "step": 2645
235
+ },
236
+ {
237
+ "epoch": 5.0,
238
+ "eval_exact_match": 25.8,
239
+ "eval_f1": 37.67571428571429,
240
+ "step": 2645
241
+ },
242
+ {
243
+ "epoch": 5.1,
244
+ "learning_rate": 3e-05,
245
+ "loss": 1.2195,
246
+ "step": 2700
247
+ },
248
+ {
249
+ "epoch": 5.29,
250
+ "learning_rate": 3e-05,
251
+ "loss": 1.2003,
252
+ "step": 2800
253
+ },
254
+ {
255
+ "epoch": 5.48,
256
+ "learning_rate": 3e-05,
257
+ "loss": 1.1864,
258
+ "step": 2900
259
+ },
260
+ {
261
+ "epoch": 5.67,
262
+ "learning_rate": 3e-05,
263
+ "loss": 1.2231,
264
+ "step": 3000
265
+ },
266
+ {
267
+ "epoch": 5.86,
268
+ "learning_rate": 3e-05,
269
+ "loss": 1.1972,
270
+ "step": 3100
271
+ },
272
+ {
273
+ "epoch": 6.0,
274
+ "eval_accuracy": 0.6125641025641025,
275
+ "eval_loss": 1.4135066270828247,
276
+ "eval_runtime": 5.0312,
277
+ "eval_samples_per_second": 99.38,
278
+ "eval_steps_per_second": 12.522,
279
+ "step": 3174
280
+ },
281
+ {
282
+ "epoch": 6.0,
283
+ "eval_exact_match": 24.0,
284
+ "eval_f1": 36.140952380952385,
285
+ "step": 3174
286
+ },
287
+ {
288
+ "epoch": 6.05,
289
+ "learning_rate": 3e-05,
290
+ "loss": 1.1889,
291
+ "step": 3200
292
+ },
293
+ {
294
+ "epoch": 6.24,
295
+ "learning_rate": 3e-05,
296
+ "loss": 1.1471,
297
+ "step": 3300
298
+ },
299
+ {
300
+ "epoch": 6.43,
301
+ "learning_rate": 3e-05,
302
+ "loss": 1.1509,
303
+ "step": 3400
304
+ },
305
+ {
306
+ "epoch": 6.62,
307
+ "learning_rate": 3e-05,
308
+ "loss": 1.1654,
309
+ "step": 3500
310
+ },
311
+ {
312
+ "epoch": 6.81,
313
+ "learning_rate": 3e-05,
314
+ "loss": 1.1716,
315
+ "step": 3600
316
+ },
317
+ {
318
+ "epoch": 6.99,
319
+ "learning_rate": 3e-05,
320
+ "loss": 1.1839,
321
+ "step": 3700
322
+ },
323
+ {
324
+ "epoch": 7.0,
325
+ "eval_accuracy": 0.6081025641025641,
326
+ "eval_loss": 1.500714898109436,
327
+ "eval_runtime": 4.4677,
328
+ "eval_samples_per_second": 111.916,
329
+ "eval_steps_per_second": 14.101,
330
+ "step": 3703
331
+ },
332
+ {
333
+ "epoch": 7.0,
334
+ "eval_exact_match": 22.8,
335
+ "eval_f1": 33.44777777777779,
336
+ "step": 3703
337
+ },
338
+ {
339
+ "epoch": 7.18,
340
+ "learning_rate": 3e-05,
341
+ "loss": 1.1059,
342
+ "step": 3800
343
+ },
344
+ {
345
+ "epoch": 7.37,
346
+ "learning_rate": 3e-05,
347
+ "loss": 1.1146,
348
+ "step": 3900
349
+ },
350
+ {
351
+ "epoch": 7.56,
352
+ "learning_rate": 3e-05,
353
+ "loss": 1.1121,
354
+ "step": 4000
355
+ },
356
+ {
357
+ "epoch": 7.75,
358
+ "learning_rate": 3e-05,
359
+ "loss": 1.1286,
360
+ "step": 4100
361
+ },
362
+ {
363
+ "epoch": 7.94,
364
+ "learning_rate": 3e-05,
365
+ "loss": 1.1334,
366
+ "step": 4200
367
+ },
368
+ {
369
+ "epoch": 8.0,
370
+ "eval_accuracy": 0.6074358974358974,
371
+ "eval_loss": 1.524243950843811,
372
+ "eval_runtime": 4.2161,
373
+ "eval_samples_per_second": 118.592,
374
+ "eval_steps_per_second": 14.943,
375
+ "step": 4232
376
+ },
377
+ {
378
+ "epoch": 8.0,
379
+ "eval_exact_match": 23.2,
380
+ "eval_f1": 34.507792207792214,
381
+ "step": 4232
382
+ },
383
+ {
384
+ "epoch": 8.13,
385
+ "learning_rate": 3e-05,
386
+ "loss": 1.1072,
387
+ "step": 4300
388
+ },
389
+ {
390
+ "epoch": 8.32,
391
+ "learning_rate": 3e-05,
392
+ "loss": 1.0626,
393
+ "step": 4400
394
+ },
395
+ {
396
+ "epoch": 8.51,
397
+ "learning_rate": 3e-05,
398
+ "loss": 1.0842,
399
+ "step": 4500
400
+ },
401
+ {
402
+ "epoch": 8.7,
403
+ "learning_rate": 3e-05,
404
+ "loss": 1.0812,
405
+ "step": 4600
406
+ },
407
+ {
408
+ "epoch": 8.88,
409
+ "learning_rate": 3e-05,
410
+ "loss": 1.0966,
411
+ "step": 4700
412
+ },
413
+ {
414
+ "epoch": 9.0,
415
+ "eval_accuracy": 0.5802564102564103,
416
+ "eval_loss": 1.6106736660003662,
417
+ "eval_runtime": 4.2194,
418
+ "eval_samples_per_second": 118.5,
419
+ "eval_steps_per_second": 14.931,
420
+ "step": 4761
421
+ },
422
+ {
423
+ "epoch": 9.0,
424
+ "eval_exact_match": 22.8,
425
+ "eval_f1": 34.45666666666666,
426
+ "step": 4761
427
+ },
428
+ {
429
+ "epoch": 9.07,
430
+ "learning_rate": 3e-05,
431
+ "loss": 1.0755,
432
+ "step": 4800
433
+ },
434
+ {
435
+ "epoch": 9.26,
436
+ "learning_rate": 3e-05,
437
+ "loss": 1.0189,
438
+ "step": 4900
439
+ },
440
+ {
441
+ "epoch": 9.45,
442
+ "learning_rate": 3e-05,
443
+ "loss": 1.0565,
444
+ "step": 5000
445
+ },
446
+ {
447
+ "epoch": 9.64,
448
+ "learning_rate": 3e-05,
449
+ "loss": 1.0635,
450
+ "step": 5100
451
+ },
452
+ {
453
+ "epoch": 9.83,
454
+ "learning_rate": 3e-05,
455
+ "loss": 1.0485,
456
+ "step": 5200
457
+ },
458
+ {
459
+ "epoch": 10.0,
460
+ "eval_accuracy": 0.6049230769230769,
461
+ "eval_loss": 1.6749216318130493,
462
+ "eval_runtime": 4.8521,
463
+ "eval_samples_per_second": 103.048,
464
+ "eval_steps_per_second": 12.984,
465
+ "step": 5290
466
+ },
467
+ {
468
+ "epoch": 10.0,
469
+ "eval_exact_match": 22.8,
470
+ "eval_f1": 34.3535064935065,
471
+ "step": 5290
472
+ },
473
+ {
474
+ "epoch": 10.02,
475
+ "learning_rate": 3e-05,
476
+ "loss": 1.0585,
477
+ "step": 5300
478
+ },
479
+ {
480
+ "epoch": 10.21,
481
+ "learning_rate": 3e-05,
482
+ "loss": 1.0098,
483
+ "step": 5400
484
+ },
485
+ {
486
+ "epoch": 10.4,
487
+ "learning_rate": 3e-05,
488
+ "loss": 0.9935,
489
+ "step": 5500
490
+ },
491
+ {
492
+ "epoch": 10.59,
493
+ "learning_rate": 3e-05,
494
+ "loss": 1.0178,
495
+ "step": 5600
496
+ },
497
+ {
498
+ "epoch": 10.78,
499
+ "learning_rate": 3e-05,
500
+ "loss": 1.0217,
501
+ "step": 5700
502
+ },
503
+ {
504
+ "epoch": 10.96,
505
+ "learning_rate": 3e-05,
506
+ "loss": 1.021,
507
+ "step": 5800
508
+ },
509
+ {
510
+ "epoch": 11.0,
511
+ "eval_accuracy": 0.6014871794871794,
512
+ "eval_loss": 1.7323671579360962,
513
+ "eval_runtime": 4.4139,
514
+ "eval_samples_per_second": 113.28,
515
+ "eval_steps_per_second": 14.273,
516
+ "step": 5819
517
+ },
518
+ {
519
+ "epoch": 11.0,
520
+ "eval_exact_match": 22.4,
521
+ "eval_f1": 34.060490620490626,
522
+ "step": 5819
523
+ },
524
+ {
525
+ "epoch": 11.15,
526
+ "learning_rate": 3e-05,
527
+ "loss": 0.9788,
528
+ "step": 5900
529
+ },
530
+ {
531
+ "epoch": 11.34,
532
+ "learning_rate": 3e-05,
533
+ "loss": 0.9799,
534
+ "step": 6000
535
+ },
536
+ {
537
+ "epoch": 11.53,
538
+ "learning_rate": 3e-05,
539
+ "loss": 0.9749,
540
+ "step": 6100
541
+ },
542
+ {
543
+ "epoch": 11.72,
544
+ "learning_rate": 3e-05,
545
+ "loss": 0.981,
546
+ "step": 6200
547
+ },
548
+ {
549
+ "epoch": 11.91,
550
+ "learning_rate": 3e-05,
551
+ "loss": 0.9918,
552
+ "step": 6300
553
+ },
554
+ {
555
+ "epoch": 12.0,
556
+ "eval_accuracy": 0.6007179487179487,
557
+ "eval_loss": 1.7632156610488892,
558
+ "eval_runtime": 4.4251,
559
+ "eval_samples_per_second": 112.992,
560
+ "eval_steps_per_second": 14.237,
561
+ "step": 6348
562
+ },
563
+ {
564
+ "epoch": 12.0,
565
+ "eval_exact_match": 22.4,
566
+ "eval_f1": 33.017142857142865,
567
+ "step": 6348
568
+ },
569
+ {
570
+ "epoch": 12.1,
571
+ "learning_rate": 3e-05,
572
+ "loss": 0.967,
573
+ "step": 6400
574
+ },
575
+ {
576
+ "epoch": 12.29,
577
+ "learning_rate": 3e-05,
578
+ "loss": 0.9341,
579
+ "step": 6500
580
+ },
581
+ {
582
+ "epoch": 12.48,
583
+ "learning_rate": 3e-05,
584
+ "loss": 0.9656,
585
+ "step": 6600
586
+ },
587
+ {
588
+ "epoch": 12.67,
589
+ "learning_rate": 3e-05,
590
+ "loss": 0.9519,
591
+ "step": 6700
592
+ },
593
+ {
594
+ "epoch": 12.85,
595
+ "learning_rate": 3e-05,
596
+ "loss": 0.947,
597
+ "step": 6800
598
+ },
599
+ {
600
+ "epoch": 13.0,
601
+ "eval_accuracy": 0.6010769230769231,
602
+ "eval_loss": 1.8302537202835083,
603
+ "eval_runtime": 4.849,
604
+ "eval_samples_per_second": 103.115,
605
+ "eval_steps_per_second": 12.992,
606
+ "step": 6877
607
+ },
608
+ {
609
+ "epoch": 13.0,
610
+ "eval_exact_match": 21.6,
611
+ "eval_f1": 32.89746031746034,
612
+ "step": 6877
613
+ },
614
+ {
615
+ "epoch": 13.04,
616
+ "learning_rate": 3e-05,
617
+ "loss": 0.9461,
618
+ "step": 6900
619
+ },
620
+ {
621
+ "epoch": 13.23,
622
+ "learning_rate": 3e-05,
623
+ "loss": 0.9153,
624
+ "step": 7000
625
+ },
626
+ {
627
+ "epoch": 13.42,
628
+ "learning_rate": 3e-05,
629
+ "loss": 0.9166,
630
+ "step": 7100
631
+ },
632
+ {
633
+ "epoch": 13.61,
634
+ "learning_rate": 3e-05,
635
+ "loss": 0.9266,
636
+ "step": 7200
637
+ },
638
+ {
639
+ "epoch": 13.8,
640
+ "learning_rate": 3e-05,
641
+ "loss": 0.9089,
642
+ "step": 7300
643
+ },
644
+ {
645
+ "epoch": 13.99,
646
+ "learning_rate": 3e-05,
647
+ "loss": 0.9376,
648
+ "step": 7400
649
+ },
650
+ {
651
+ "epoch": 14.0,
652
+ "eval_accuracy": 0.5990769230769231,
653
+ "eval_loss": 1.8872731924057007,
654
+ "eval_runtime": 4.1087,
655
+ "eval_samples_per_second": 121.694,
656
+ "eval_steps_per_second": 15.333,
657
+ "step": 7406
658
+ },
659
+ {
660
+ "epoch": 14.0,
661
+ "eval_exact_match": 20.8,
662
+ "eval_f1": 32.45952380952382,
663
+ "step": 7406
664
+ },
665
+ {
666
+ "epoch": 14.18,
667
+ "learning_rate": 3e-05,
668
+ "loss": 0.8792,
669
+ "step": 7500
670
+ },
671
+ {
672
+ "epoch": 14.37,
673
+ "learning_rate": 3e-05,
674
+ "loss": 0.8787,
675
+ "step": 7600
676
+ },
677
+ {
678
+ "epoch": 14.56,
679
+ "learning_rate": 3e-05,
680
+ "loss": 0.874,
681
+ "step": 7700
682
+ },
683
+ {
684
+ "epoch": 14.74,
685
+ "learning_rate": 3e-05,
686
+ "loss": 0.9123,
687
+ "step": 7800
688
+ },
689
+ {
690
+ "epoch": 14.93,
691
+ "learning_rate": 3e-05,
692
+ "loss": 0.898,
693
+ "step": 7900
694
+ },
695
+ {
696
+ "epoch": 15.0,
697
+ "eval_accuracy": 0.5975897435897436,
698
+ "eval_loss": 1.96884024143219,
699
+ "eval_runtime": 4.5009,
700
+ "eval_samples_per_second": 111.09,
701
+ "eval_steps_per_second": 13.997,
702
+ "step": 7935
703
+ },
704
+ {
705
+ "epoch": 15.0,
706
+ "eval_exact_match": 20.6,
707
+ "eval_f1": 31.090952380952388,
708
+ "step": 7935
709
+ },
710
+ {
711
+ "epoch": 15.12,
712
+ "learning_rate": 3e-05,
713
+ "loss": 0.8631,
714
+ "step": 8000
715
+ },
716
+ {
717
+ "epoch": 15.31,
718
+ "learning_rate": 3e-05,
719
+ "loss": 0.8353,
720
+ "step": 8100
721
+ },
722
+ {
723
+ "epoch": 15.5,
724
+ "learning_rate": 3e-05,
725
+ "loss": 0.8673,
726
+ "step": 8200
727
+ },
728
+ {
729
+ "epoch": 15.69,
730
+ "learning_rate": 3e-05,
731
+ "loss": 0.8515,
732
+ "step": 8300
733
+ },
734
+ {
735
+ "epoch": 15.88,
736
+ "learning_rate": 3e-05,
737
+ "loss": 0.8559,
738
+ "step": 8400
739
+ },
740
+ {
741
+ "epoch": 16.0,
742
+ "eval_accuracy": 0.5987692307692307,
743
+ "eval_loss": 1.9724328517913818,
744
+ "eval_runtime": 4.5341,
745
+ "eval_samples_per_second": 110.275,
746
+ "eval_steps_per_second": 13.895,
747
+ "step": 8464
748
+ },
749
+ {
750
+ "epoch": 16.0,
751
+ "eval_exact_match": 21.4,
752
+ "eval_f1": 32.21255411255413,
753
+ "step": 8464
754
+ },
755
+ {
756
+ "epoch": 16.07,
757
+ "learning_rate": 3e-05,
758
+ "loss": 0.8508,
759
+ "step": 8500
760
+ },
761
+ {
762
+ "epoch": 16.26,
763
+ "learning_rate": 3e-05,
764
+ "loss": 0.8122,
765
+ "step": 8600
766
+ },
767
+ {
768
+ "epoch": 16.45,
769
+ "learning_rate": 3e-05,
770
+ "loss": 0.8223,
771
+ "step": 8700
772
+ },
773
+ {
774
+ "epoch": 16.64,
775
+ "learning_rate": 3e-05,
776
+ "loss": 0.8275,
777
+ "step": 8800
778
+ },
779
+ {
780
+ "epoch": 16.82,
781
+ "learning_rate": 3e-05,
782
+ "loss": 0.8348,
783
+ "step": 8900
784
+ },
785
+ {
786
+ "epoch": 17.0,
787
+ "eval_accuracy": 0.5713846153846154,
788
+ "eval_loss": 1.9815243482589722,
789
+ "eval_runtime": 5.0655,
790
+ "eval_samples_per_second": 98.706,
791
+ "eval_steps_per_second": 12.437,
792
+ "step": 8993
793
+ },
794
+ {
795
+ "epoch": 17.0,
796
+ "eval_exact_match": 20.0,
797
+ "eval_f1": 30.670173160173174,
798
+ "step": 8993
799
+ },
800
+ {
801
+ "epoch": 17.01,
802
+ "learning_rate": 3e-05,
803
+ "loss": 0.837,
804
+ "step": 9000
805
+ },
806
+ {
807
+ "epoch": 17.2,
808
+ "learning_rate": 3e-05,
809
+ "loss": 0.7764,
810
+ "step": 9100
811
+ },
812
+ {
813
+ "epoch": 17.39,
814
+ "learning_rate": 3e-05,
815
+ "loss": 0.7817,
816
+ "step": 9200
817
+ },
818
+ {
819
+ "epoch": 17.58,
820
+ "learning_rate": 3e-05,
821
+ "loss": 0.8068,
822
+ "step": 9300
823
+ },
824
+ {
825
+ "epoch": 17.77,
826
+ "learning_rate": 3e-05,
827
+ "loss": 0.7968,
828
+ "step": 9400
829
+ },
830
+ {
831
+ "epoch": 17.96,
832
+ "learning_rate": 3e-05,
833
+ "loss": 0.8106,
834
+ "step": 9500
835
+ },
836
+ {
837
+ "epoch": 18.0,
838
+ "eval_accuracy": 0.598,
839
+ "eval_loss": 2.038572072982788,
840
+ "eval_runtime": 4.2168,
841
+ "eval_samples_per_second": 118.574,
842
+ "eval_steps_per_second": 14.94,
843
+ "step": 9522
844
+ },
845
+ {
846
+ "epoch": 18.0,
847
+ "eval_exact_match": 19.2,
848
+ "eval_f1": 30.44984126984128,
849
+ "step": 9522
850
+ },
851
+ {
852
+ "epoch": 18.15,
853
+ "learning_rate": 3e-05,
854
+ "loss": 0.7697,
855
+ "step": 9600
856
+ },
857
+ {
858
+ "epoch": 18.34,
859
+ "learning_rate": 3e-05,
860
+ "loss": 0.7429,
861
+ "step": 9700
862
+ },
863
+ {
864
+ "epoch": 18.53,
865
+ "learning_rate": 3e-05,
866
+ "loss": 0.7468,
867
+ "step": 9800
868
+ },
869
+ {
870
+ "epoch": 18.71,
871
+ "learning_rate": 3e-05,
872
+ "loss": 0.7667,
873
+ "step": 9900
874
+ },
875
+ {
876
+ "epoch": 18.9,
877
+ "learning_rate": 3e-05,
878
+ "loss": 0.7848,
879
+ "step": 10000
880
+ },
881
+ {
882
+ "epoch": 19.0,
883
+ "eval_accuracy": 0.5963589743589743,
884
+ "eval_loss": 2.0627450942993164,
885
+ "eval_runtime": 4.2486,
886
+ "eval_samples_per_second": 117.685,
887
+ "eval_steps_per_second": 14.828,
888
+ "step": 10051
889
+ },
890
+ {
891
+ "epoch": 19.0,
892
+ "eval_exact_match": 19.0,
893
+ "eval_f1": 29.635238095238098,
894
+ "step": 10051
895
+ },
896
+ {
897
+ "epoch": 19.09,
898
+ "learning_rate": 3e-05,
899
+ "loss": 0.7537,
900
+ "step": 10100
901
+ },
902
+ {
903
+ "epoch": 19.28,
904
+ "learning_rate": 3e-05,
905
+ "loss": 0.718,
906
+ "step": 10200
907
+ },
908
+ {
909
+ "epoch": 19.47,
910
+ "learning_rate": 3e-05,
911
+ "loss": 0.7241,
912
+ "step": 10300
913
+ },
914
+ {
915
+ "epoch": 19.66,
916
+ "learning_rate": 3e-05,
917
+ "loss": 0.7421,
918
+ "step": 10400
919
+ },
920
+ {
921
+ "epoch": 19.85,
922
+ "learning_rate": 3e-05,
923
+ "loss": 0.745,
924
+ "step": 10500
925
+ },
926
+ {
927
+ "epoch": 20.0,
928
+ "eval_accuracy": 0.5965641025641025,
929
+ "eval_loss": 2.0825321674346924,
930
+ "eval_runtime": 4.9023,
931
+ "eval_samples_per_second": 101.993,
932
+ "eval_steps_per_second": 12.851,
933
+ "step": 10580
934
+ },
935
+ {
936
+ "epoch": 20.0,
937
+ "eval_exact_match": 16.0,
938
+ "eval_f1": 25.784563976747563,
939
+ "step": 10580
940
+ },
941
+ {
942
+ "epoch": 20.04,
943
+ "learning_rate": 3e-05,
944
+ "loss": 0.7384,
945
+ "step": 10600
946
+ },
947
+ {
948
+ "epoch": 20.23,
949
+ "learning_rate": 3e-05,
950
+ "loss": 0.6786,
951
+ "step": 10700
952
+ },
953
+ {
954
+ "epoch": 20.42,
955
+ "learning_rate": 3e-05,
956
+ "loss": 0.6921,
957
+ "step": 10800
958
+ },
959
+ {
960
+ "epoch": 20.6,
961
+ "learning_rate": 3e-05,
962
+ "loss": 0.7092,
963
+ "step": 10900
964
+ },
965
+ {
966
+ "epoch": 20.79,
967
+ "learning_rate": 3e-05,
968
+ "loss": 0.7248,
969
+ "step": 11000
970
+ },
971
+ {
972
+ "epoch": 20.98,
973
+ "learning_rate": 3e-05,
974
+ "loss": 0.7208,
975
+ "step": 11100
976
+ },
977
+ {
978
+ "epoch": 21.0,
979
+ "eval_accuracy": 0.595948717948718,
980
+ "eval_loss": 2.0958657264709473,
981
+ "eval_runtime": 4.2167,
982
+ "eval_samples_per_second": 118.575,
983
+ "eval_steps_per_second": 14.941,
984
+ "step": 11109
985
+ },
986
+ {
987
+ "epoch": 21.0,
988
+ "eval_exact_match": 9.4,
989
+ "eval_f1": 15.48481664400564,
990
+ "step": 11109
991
+ },
992
+ {
993
+ "epoch": 21.17,
994
+ "learning_rate": 3e-05,
995
+ "loss": 0.6604,
996
+ "step": 11200
997
+ },
998
+ {
999
+ "epoch": 21.36,
1000
+ "learning_rate": 3e-05,
1001
+ "loss": 0.6599,
1002
+ "step": 11300
1003
+ },
1004
+ {
1005
+ "epoch": 21.55,
1006
+ "learning_rate": 3e-05,
1007
+ "loss": 0.6775,
1008
+ "step": 11400
1009
+ },
1010
+ {
1011
+ "epoch": 21.74,
1012
+ "learning_rate": 3e-05,
1013
+ "loss": 0.7001,
1014
+ "step": 11500
1015
+ },
1016
+ {
1017
+ "epoch": 21.93,
1018
+ "learning_rate": 3e-05,
1019
+ "loss": 0.6842,
1020
+ "step": 11600
1021
+ },
1022
+ {
1023
+ "epoch": 22.0,
1024
+ "eval_accuracy": 0.5967692307692307,
1025
+ "eval_loss": 2.153414487838745,
1026
+ "eval_runtime": 4.8531,
1027
+ "eval_samples_per_second": 103.026,
1028
+ "eval_steps_per_second": 12.981,
1029
+ "step": 11638
1030
+ },
1031
+ {
1032
+ "epoch": 22.0,
1033
+ "eval_exact_match": 9.2,
1034
+ "eval_f1": 17.03231000578938,
1035
+ "step": 11638
1036
+ },
1037
+ {
1038
+ "epoch": 22.12,
1039
+ "learning_rate": 3e-05,
1040
+ "loss": 0.6544,
1041
+ "step": 11700
1042
+ },
1043
+ {
1044
+ "epoch": 22.31,
1045
+ "learning_rate": 3e-05,
1046
+ "loss": 0.641,
1047
+ "step": 11800
1048
+ },
1049
+ {
1050
+ "epoch": 22.5,
1051
+ "learning_rate": 3e-05,
1052
+ "loss": 0.6369,
1053
+ "step": 11900
1054
+ },
1055
+ {
1056
+ "epoch": 22.68,
1057
+ "learning_rate": 3e-05,
1058
+ "loss": 0.6468,
1059
+ "step": 12000
1060
+ },
1061
+ {
1062
+ "epoch": 22.87,
1063
+ "learning_rate": 3e-05,
1064
+ "loss": 0.6661,
1065
+ "step": 12100
1066
+ },
1067
+ {
1068
+ "epoch": 23.0,
1069
+ "eval_accuracy": 0.5974871794871794,
1070
+ "eval_loss": 2.179178237915039,
1071
+ "eval_runtime": 4.8444,
1072
+ "eval_samples_per_second": 103.211,
1073
+ "eval_steps_per_second": 13.005,
1074
+ "step": 12167
1075
+ },
1076
+ {
1077
+ "epoch": 23.0,
1078
+ "eval_exact_match": 3.6,
1079
+ "eval_f1": 7.6552922676511725,
1080
+ "step": 12167
1081
+ },
1082
+ {
1083
+ "epoch": 23.06,
1084
+ "learning_rate": 3e-05,
1085
+ "loss": 0.6418,
1086
+ "step": 12200
1087
+ },
1088
+ {
1089
+ "epoch": 23.25,
1090
+ "learning_rate": 3e-05,
1091
+ "loss": 0.6012,
1092
+ "step": 12300
1093
+ },
1094
+ {
1095
+ "epoch": 23.44,
1096
+ "learning_rate": 3e-05,
1097
+ "loss": 0.6211,
1098
+ "step": 12400
1099
+ },
1100
+ {
1101
+ "epoch": 23.63,
1102
+ "learning_rate": 3e-05,
1103
+ "loss": 0.6174,
1104
+ "step": 12500
1105
+ },
1106
+ {
1107
+ "epoch": 23.82,
1108
+ "learning_rate": 3e-05,
1109
+ "loss": 0.6193,
1110
+ "step": 12600
1111
+ },
1112
+ {
1113
+ "epoch": 24.0,
1114
+ "eval_accuracy": 0.5967179487179487,
1115
+ "eval_loss": 2.152991771697998,
1116
+ "eval_runtime": 4.429,
1117
+ "eval_samples_per_second": 112.891,
1118
+ "eval_steps_per_second": 14.224,
1119
+ "step": 12696
1120
+ },
1121
+ {
1122
+ "epoch": 24.0,
1123
+ "eval_exact_match": 3.2,
1124
+ "eval_f1": 7.263414468771945,
1125
+ "step": 12696
1126
+ },
1127
+ {
1128
+ "epoch": 24.01,
1129
+ "learning_rate": 3e-05,
1130
+ "loss": 0.6329,
1131
+ "step": 12700
1132
+ },
1133
+ {
1134
+ "epoch": 24.2,
1135
+ "learning_rate": 3e-05,
1136
+ "loss": 0.572,
1137
+ "step": 12800
1138
+ },
1139
+ {
1140
+ "epoch": 24.39,
1141
+ "learning_rate": 3e-05,
1142
+ "loss": 0.585,
1143
+ "step": 12900
1144
+ },
1145
+ {
1146
+ "epoch": 24.57,
1147
+ "learning_rate": 3e-05,
1148
+ "loss": 0.5894,
1149
+ "step": 13000
1150
+ },
1151
+ {
1152
+ "epoch": 24.76,
1153
+ "learning_rate": 3e-05,
1154
+ "loss": 0.6002,
1155
+ "step": 13100
1156
+ },
1157
+ {
1158
+ "epoch": 24.95,
1159
+ "learning_rate": 3e-05,
1160
+ "loss": 0.6064,
1161
+ "step": 13200
1162
+ },
1163
+ {
1164
+ "epoch": 25.0,
1165
+ "eval_accuracy": 0.5957948717948718,
1166
+ "eval_loss": 2.1720352172851562,
1167
+ "eval_runtime": 4.1244,
1168
+ "eval_samples_per_second": 121.23,
1169
+ "eval_steps_per_second": 15.275,
1170
+ "step": 13225
1171
+ },
1172
+ {
1173
+ "epoch": 25.0,
1174
+ "eval_exact_match": 1.0,
1175
+ "eval_f1": 2.817526608207056,
1176
+ "step": 13225
1177
+ },
1178
+ {
1179
+ "epoch": 25.14,
1180
+ "learning_rate": 3e-05,
1181
+ "loss": 0.5603,
1182
+ "step": 13300
1183
+ },
1184
+ {
1185
+ "epoch": 25.33,
1186
+ "learning_rate": 3e-05,
1187
+ "loss": 0.5438,
1188
+ "step": 13400
1189
+ },
1190
+ {
1191
+ "epoch": 25.52,
1192
+ "learning_rate": 3e-05,
1193
+ "loss": 0.5718,
1194
+ "step": 13500
1195
+ },
1196
+ {
1197
+ "epoch": 25.71,
1198
+ "learning_rate": 3e-05,
1199
+ "loss": 0.5804,
1200
+ "step": 13600
1201
+ },
1202
+ {
1203
+ "epoch": 25.9,
1204
+ "learning_rate": 3e-05,
1205
+ "loss": 0.5776,
1206
+ "step": 13700
1207
+ },
1208
+ {
1209
+ "epoch": 26.0,
1210
+ "eval_accuracy": 0.5965641025641025,
1211
+ "eval_loss": 2.216242790222168,
1212
+ "eval_runtime": 4.8581,
1213
+ "eval_samples_per_second": 102.921,
1214
+ "eval_steps_per_second": 12.968,
1215
+ "step": 13754
1216
+ },
1217
+ {
1218
+ "epoch": 26.0,
1219
+ "eval_exact_match": 0.6,
1220
+ "eval_f1": 1.5737055770144102,
1221
+ "step": 13754
1222
+ },
1223
+ {
1224
+ "epoch": 26.09,
1225
+ "learning_rate": 3e-05,
1226
+ "loss": 0.5613,
1227
+ "step": 13800
1228
+ },
1229
+ {
1230
+ "epoch": 26.28,
1231
+ "learning_rate": 3e-05,
1232
+ "loss": 0.5267,
1233
+ "step": 13900
1234
+ },
1235
+ {
1236
+ "epoch": 26.47,
1237
+ "learning_rate": 3e-05,
1238
+ "loss": 0.542,
1239
+ "step": 14000
1240
+ },
1241
+ {
1242
+ "epoch": 26.65,
1243
+ "learning_rate": 3e-05,
1244
+ "loss": 0.5457,
1245
+ "step": 14100
1246
+ },
1247
+ {
1248
+ "epoch": 26.84,
1249
+ "learning_rate": 3e-05,
1250
+ "loss": 0.5492,
1251
+ "step": 14200
1252
+ },
1253
+ {
1254
+ "epoch": 27.0,
1255
+ "eval_accuracy": 0.5861538461538461,
1256
+ "eval_loss": 2.238172769546509,
1257
+ "eval_runtime": 4.6504,
1258
+ "eval_samples_per_second": 107.516,
1259
+ "eval_steps_per_second": 13.547,
1260
+ "step": 14283
1261
+ },
1262
+ {
1263
+ "epoch": 27.0,
1264
+ "eval_exact_match": 0.4,
1265
+ "eval_f1": 1.9589310673272307,
1266
+ "step": 14283
1267
+ },
1268
+ {
1269
+ "epoch": 27.03,
1270
+ "learning_rate": 3e-05,
1271
+ "loss": 0.5488,
1272
+ "step": 14300
1273
+ },
1274
+ {
1275
+ "epoch": 27.22,
1276
+ "learning_rate": 3e-05,
1277
+ "loss": 0.5016,
1278
+ "step": 14400
1279
+ },
1280
+ {
1281
+ "epoch": 27.41,
1282
+ "learning_rate": 3e-05,
1283
+ "loss": 0.5115,
1284
+ "step": 14500
1285
+ },
1286
+ {
1287
+ "epoch": 27.6,
1288
+ "learning_rate": 3e-05,
1289
+ "loss": 0.5219,
1290
+ "step": 14600
1291
+ },
1292
+ {
1293
+ "epoch": 27.79,
1294
+ "learning_rate": 3e-05,
1295
+ "loss": 0.5388,
1296
+ "step": 14700
1297
+ },
1298
+ {
1299
+ "epoch": 27.98,
1300
+ "learning_rate": 3e-05,
1301
+ "loss": 0.5256,
1302
+ "step": 14800
1303
+ },
1304
+ {
1305
+ "epoch": 28.0,
1306
+ "eval_accuracy": 0.5963076923076923,
1307
+ "eval_loss": 2.2273213863372803,
1308
+ "eval_runtime": 4.4195,
1309
+ "eval_samples_per_second": 113.136,
1310
+ "eval_steps_per_second": 14.255,
1311
+ "step": 14812
1312
+ },
1313
+ {
1314
+ "epoch": 28.0,
1315
+ "eval_exact_match": 0.6,
1316
+ "eval_f1": 1.254701854861382,
1317
+ "step": 14812
1318
+ },
1319
+ {
1320
+ "epoch": 28.17,
1321
+ "learning_rate": 3e-05,
1322
+ "loss": 0.4814,
1323
+ "step": 14900
1324
+ },
1325
+ {
1326
+ "epoch": 28.36,
1327
+ "learning_rate": 3e-05,
1328
+ "loss": 0.4943,
1329
+ "step": 15000
1330
+ },
1331
+ {
1332
+ "epoch": 28.54,
1333
+ "learning_rate": 3e-05,
1334
+ "loss": 0.4899,
1335
+ "step": 15100
1336
+ },
1337
+ {
1338
+ "epoch": 28.73,
1339
+ "learning_rate": 3e-05,
1340
+ "loss": 0.5047,
1341
+ "step": 15200
1342
+ },
1343
+ {
1344
+ "epoch": 28.92,
1345
+ "learning_rate": 3e-05,
1346
+ "loss": 0.5128,
1347
+ "step": 15300
1348
+ },
1349
+ {
1350
+ "epoch": 29.0,
1351
+ "eval_accuracy": 0.5947692307692307,
1352
+ "eval_loss": 2.244802713394165,
1353
+ "eval_runtime": 4.2174,
1354
+ "eval_samples_per_second": 118.556,
1355
+ "eval_steps_per_second": 14.938,
1356
+ "step": 15341
1357
+ },
1358
+ {
1359
+ "epoch": 29.0,
1360
+ "eval_exact_match": 0.0,
1361
+ "eval_f1": 0.3342622661004701,
1362
+ "step": 15341
1363
+ },
1364
+ {
1365
+ "epoch": 29.11,
1366
+ "learning_rate": 3e-05,
1367
+ "loss": 0.4776,
1368
+ "step": 15400
1369
+ },
1370
+ {
1371
+ "epoch": 29.3,
1372
+ "learning_rate": 3e-05,
1373
+ "loss": 0.4659,
1374
+ "step": 15500
1375
+ },
1376
+ {
1377
+ "epoch": 29.49,
1378
+ "learning_rate": 3e-05,
1379
+ "loss": 0.4738,
1380
+ "step": 15600
1381
+ },
1382
+ {
1383
+ "epoch": 29.68,
1384
+ "learning_rate": 3e-05,
1385
+ "loss": 0.4833,
1386
+ "step": 15700
1387
+ },
1388
+ {
1389
+ "epoch": 29.87,
1390
+ "learning_rate": 3e-05,
1391
+ "loss": 0.4846,
1392
+ "step": 15800
1393
+ },
1394
+ {
1395
+ "epoch": 30.0,
1396
+ "eval_accuracy": 0.5846153846153846,
1397
+ "eval_loss": 2.2697417736053467,
1398
+ "eval_runtime": 4.1128,
1399
+ "eval_samples_per_second": 121.571,
1400
+ "eval_steps_per_second": 15.318,
1401
+ "step": 15870
1402
+ },
1403
+ {
1404
+ "epoch": 30.0,
1405
+ "eval_exact_match": 0.0,
1406
+ "eval_f1": 0.17287697257712523,
1407
+ "step": 15870
1408
+ },
1409
+ {
1410
+ "epoch": 30.06,
1411
+ "learning_rate": 3e-05,
1412
+ "loss": 0.4719,
1413
+ "step": 15900
1414
+ },
1415
+ {
1416
+ "epoch": 30.25,
1417
+ "learning_rate": 3e-05,
1418
+ "loss": 0.4404,
1419
+ "step": 16000
1420
+ },
1421
+ {
1422
+ "epoch": 30.43,
1423
+ "learning_rate": 3e-05,
1424
+ "loss": 0.4533,
1425
+ "step": 16100
1426
+ },
1427
+ {
1428
+ "epoch": 30.62,
1429
+ "learning_rate": 3e-05,
1430
+ "loss": 0.4623,
1431
+ "step": 16200
1432
+ },
1433
+ {
1434
+ "epoch": 30.81,
1435
+ "learning_rate": 3e-05,
1436
+ "loss": 0.4623,
1437
+ "step": 16300
1438
+ },
1439
+ {
1440
+ "epoch": 31.0,
1441
+ "eval_accuracy": 0.5968205128205128,
1442
+ "eval_loss": 2.242527723312378,
1443
+ "eval_runtime": 4.221,
1444
+ "eval_samples_per_second": 118.456,
1445
+ "eval_steps_per_second": 14.925,
1446
+ "step": 16399
1447
+ },
1448
+ {
1449
+ "epoch": 31.0,
1450
+ "eval_exact_match": 0.0,
1451
+ "eval_f1": 0.03156586531586532,
1452
+ "step": 16399
1453
+ },
1454
+ {
1455
+ "epoch": 31.0,
1456
+ "learning_rate": 3e-05,
1457
+ "loss": 0.4683,
1458
+ "step": 16400
1459
+ },
1460
+ {
1461
+ "epoch": 31.19,
1462
+ "learning_rate": 3e-05,
1463
+ "loss": 0.4273,
1464
+ "step": 16500
1465
+ },
1466
+ {
1467
+ "epoch": 31.38,
1468
+ "learning_rate": 3e-05,
1469
+ "loss": 0.4315,
1470
+ "step": 16600
1471
+ },
1472
+ {
1473
+ "epoch": 31.57,
1474
+ "learning_rate": 3e-05,
1475
+ "loss": 0.4442,
1476
+ "step": 16700
1477
+ },
1478
+ {
1479
+ "epoch": 31.76,
1480
+ "learning_rate": 3e-05,
1481
+ "loss": 0.4539,
1482
+ "step": 16800
1483
+ },
1484
+ {
1485
+ "epoch": 31.95,
1486
+ "learning_rate": 3e-05,
1487
+ "loss": 0.4468,
1488
+ "step": 16900
1489
+ },
1490
+ {
1491
+ "epoch": 32.0,
1492
+ "eval_accuracy": 0.5936410256410256,
1493
+ "eval_loss": 2.265359401702881,
1494
+ "eval_runtime": 4.2139,
1495
+ "eval_samples_per_second": 118.656,
1496
+ "eval_steps_per_second": 14.951,
1497
+ "step": 16928
1498
+ },
1499
+ {
1500
+ "epoch": 32.0,
1501
+ "eval_exact_match": 0.0,
1502
+ "eval_f1": 0.0,
1503
+ "step": 16928
1504
+ },
1505
+ {
1506
+ "epoch": 32.14,
1507
+ "learning_rate": 3e-05,
1508
+ "loss": 0.4321,
1509
+ "step": 17000
1510
+ },
1511
+ {
1512
+ "epoch": 32.33,
1513
+ "learning_rate": 3e-05,
1514
+ "loss": 0.4213,
1515
+ "step": 17100
1516
+ },
1517
+ {
1518
+ "epoch": 32.51,
1519
+ "learning_rate": 3e-05,
1520
+ "loss": 0.4312,
1521
+ "step": 17200
1522
+ },
1523
+ {
1524
+ "epoch": 32.7,
1525
+ "learning_rate": 3e-05,
1526
+ "loss": 0.442,
1527
+ "step": 17300
1528
+ },
1529
+ {
1530
+ "epoch": 32.89,
1531
+ "learning_rate": 3e-05,
1532
+ "loss": 0.4714,
1533
+ "step": 17400
1534
+ },
1535
+ {
1536
+ "epoch": 33.0,
1537
+ "eval_accuracy": 0.5956923076923077,
1538
+ "eval_loss": 2.1317203044891357,
1539
+ "eval_runtime": 4.6149,
1540
+ "eval_samples_per_second": 108.344,
1541
+ "eval_steps_per_second": 13.651,
1542
+ "step": 17457
1543
+ },
1544
+ {
1545
+ "epoch": 33.0,
1546
+ "eval_exact_match": 0.0,
1547
+ "eval_f1": 0.0,
1548
+ "step": 17457
1549
+ },
1550
+ {
1551
+ "epoch": 33.08,
1552
+ "learning_rate": 3e-05,
1553
+ "loss": 0.913,
1554
+ "step": 17500
1555
+ },
1556
+ {
1557
+ "epoch": 33.27,
1558
+ "learning_rate": 3e-05,
1559
+ "loss": 1.1341,
1560
+ "step": 17600
1561
+ },
1562
+ {
1563
+ "epoch": 33.46,
1564
+ "learning_rate": 3e-05,
1565
+ "loss": 0.9858,
1566
+ "step": 17700
1567
+ },
1568
+ {
1569
+ "epoch": 33.65,
1570
+ "learning_rate": 3e-05,
1571
+ "loss": 0.8735,
1572
+ "step": 17800
1573
+ },
1574
+ {
1575
+ "epoch": 33.84,
1576
+ "learning_rate": 3e-05,
1577
+ "loss": 0.8308,
1578
+ "step": 17900
1579
+ },
1580
+ {
1581
+ "epoch": 34.0,
1582
+ "eval_accuracy": 0.5973333333333334,
1583
+ "eval_loss": 1.9391597509384155,
1584
+ "eval_runtime": 4.8479,
1585
+ "eval_samples_per_second": 103.137,
1586
+ "eval_steps_per_second": 12.995,
1587
+ "step": 17986
1588
+ },
1589
+ {
1590
+ "epoch": 34.0,
1591
+ "eval_exact_match": 0.0,
1592
+ "eval_f1": 0.0,
1593
+ "step": 17986
1594
+ },
1595
+ {
1596
+ "epoch": 34.03,
1597
+ "learning_rate": 3e-05,
1598
+ "loss": 0.789,
1599
+ "step": 18000
1600
+ },
1601
+ {
1602
+ "epoch": 34.22,
1603
+ "learning_rate": 3e-05,
1604
+ "loss": 0.7127,
1605
+ "step": 18100
1606
+ },
1607
+ {
1608
+ "epoch": 34.4,
1609
+ "learning_rate": 3e-05,
1610
+ "loss": 0.7207,
1611
+ "step": 18200
1612
+ },
1613
+ {
1614
+ "epoch": 34.59,
1615
+ "learning_rate": 3e-05,
1616
+ "loss": 0.6766,
1617
+ "step": 18300
1618
+ },
1619
+ {
1620
+ "epoch": 34.78,
1621
+ "learning_rate": 3e-05,
1622
+ "loss": 0.6589,
1623
+ "step": 18400
1624
+ },
1625
+ {
1626
+ "epoch": 34.97,
1627
+ "learning_rate": 3e-05,
1628
+ "loss": 0.6478,
1629
+ "step": 18500
1630
+ },
1631
+ {
1632
+ "epoch": 35.0,
1633
+ "eval_accuracy": 0.5978974358974359,
1634
+ "eval_loss": 2.034553289413452,
1635
+ "eval_runtime": 4.1113,
1636
+ "eval_samples_per_second": 121.615,
1637
+ "eval_steps_per_second": 15.324,
1638
+ "step": 18515
1639
+ },
1640
+ {
1641
+ "epoch": 35.0,
1642
+ "eval_exact_match": 0.0,
1643
+ "eval_f1": 0.0,
1644
+ "step": 18515
1645
+ },
1646
+ {
1647
+ "epoch": 35.16,
1648
+ "learning_rate": 3e-05,
1649
+ "loss": 0.6129,
1650
+ "step": 18600
1651
+ },
1652
+ {
1653
+ "epoch": 35.35,
1654
+ "learning_rate": 3e-05,
1655
+ "loss": 0.6141,
1656
+ "step": 18700
1657
+ },
1658
+ {
1659
+ "epoch": 35.54,
1660
+ "learning_rate": 3e-05,
1661
+ "loss": 0.605,
1662
+ "step": 18800
1663
+ },
1664
+ {
1665
+ "epoch": 35.73,
1666
+ "learning_rate": 3e-05,
1667
+ "loss": 0.6031,
1668
+ "step": 18900
1669
+ },
1670
+ {
1671
+ "epoch": 35.92,
1672
+ "learning_rate": 3e-05,
1673
+ "loss": 0.612,
1674
+ "step": 19000
1675
+ },
1676
+ {
1677
+ "epoch": 36.0,
1678
+ "eval_accuracy": 0.5977948717948718,
1679
+ "eval_loss": 2.08821964263916,
1680
+ "eval_runtime": 4.4206,
1681
+ "eval_samples_per_second": 113.108,
1682
+ "eval_steps_per_second": 14.252,
1683
+ "step": 19044
1684
+ },
1685
+ {
1686
+ "epoch": 36.0,
1687
+ "eval_exact_match": 0.0,
1688
+ "eval_f1": 0.0,
1689
+ "step": 19044
1690
+ },
1691
+ {
1692
+ "epoch": 36.11,
1693
+ "learning_rate": 3e-05,
1694
+ "loss": 0.6204,
1695
+ "step": 19100
1696
+ },
1697
+ {
1698
+ "epoch": 36.29,
1699
+ "learning_rate": 3e-05,
1700
+ "loss": 0.569,
1701
+ "step": 19200
1702
+ },
1703
+ {
1704
+ "epoch": 36.48,
1705
+ "learning_rate": 3e-05,
1706
+ "loss": 0.5727,
1707
+ "step": 19300
1708
+ },
1709
+ {
1710
+ "epoch": 36.67,
1711
+ "learning_rate": 3e-05,
1712
+ "loss": 0.6022,
1713
+ "step": 19400
1714
+ },
1715
+ {
1716
+ "epoch": 36.86,
1717
+ "learning_rate": 3e-05,
1718
+ "loss": 0.5928,
1719
+ "step": 19500
1720
+ },
1721
+ {
1722
+ "epoch": 37.0,
1723
+ "eval_accuracy": 0.596974358974359,
1724
+ "eval_loss": 2.142036199569702,
1725
+ "eval_runtime": 4.1124,
1726
+ "eval_samples_per_second": 121.585,
1727
+ "eval_steps_per_second": 15.32,
1728
+ "step": 19573
1729
+ },
1730
+ {
1731
+ "epoch": 37.0,
1732
+ "eval_exact_match": 0.0,
1733
+ "eval_f1": 0.0,
1734
+ "step": 19573
1735
+ },
1736
+ {
1737
+ "epoch": 37.05,
1738
+ "learning_rate": 3e-05,
1739
+ "loss": 0.5796,
1740
+ "step": 19600
1741
+ },
1742
+ {
1743
+ "epoch": 37.24,
1744
+ "learning_rate": 3e-05,
1745
+ "loss": 0.5569,
1746
+ "step": 19700
1747
+ },
1748
+ {
1749
+ "epoch": 37.43,
1750
+ "learning_rate": 3e-05,
1751
+ "loss": 0.5552,
1752
+ "step": 19800
1753
+ },
1754
+ {
1755
+ "epoch": 37.62,
1756
+ "learning_rate": 3e-05,
1757
+ "loss": 0.5612,
1758
+ "step": 19900
1759
+ },
1760
+ {
1761
+ "epoch": 37.81,
1762
+ "learning_rate": 3e-05,
1763
+ "loss": 0.5765,
1764
+ "step": 20000
1765
+ },
1766
+ {
1767
+ "epoch": 38.0,
1768
+ "learning_rate": 3e-05,
1769
+ "loss": 0.5698,
1770
+ "step": 20100
1771
+ },
1772
+ {
1773
+ "epoch": 38.0,
1774
+ "eval_accuracy": 0.5966153846153847,
1775
+ "eval_loss": 2.156949758529663,
1776
+ "eval_runtime": 5.1401,
1777
+ "eval_samples_per_second": 97.274,
1778
+ "eval_steps_per_second": 12.257,
1779
+ "step": 20102
1780
+ },
1781
+ {
1782
+ "epoch": 38.0,
1783
+ "eval_exact_match": 0.0,
1784
+ "eval_f1": 0.0,
1785
+ "step": 20102
1786
+ },
1787
+ {
1788
+ "epoch": 38.19,
1789
+ "learning_rate": 3e-05,
1790
+ "loss": 0.5182,
1791
+ "step": 20200
1792
+ },
1793
+ {
1794
+ "epoch": 38.37,
1795
+ "learning_rate": 3e-05,
1796
+ "loss": 0.5277,
1797
+ "step": 20300
1798
+ },
1799
+ {
1800
+ "epoch": 38.56,
1801
+ "learning_rate": 3e-05,
1802
+ "loss": 0.5312,
1803
+ "step": 20400
1804
+ },
1805
+ {
1806
+ "epoch": 38.75,
1807
+ "learning_rate": 3e-05,
1808
+ "loss": 0.5496,
1809
+ "step": 20500
1810
+ },
1811
+ {
1812
+ "epoch": 38.94,
1813
+ "learning_rate": 3e-05,
1814
+ "loss": 0.5444,
1815
+ "step": 20600
1816
+ },
1817
+ {
1818
+ "epoch": 39.0,
1819
+ "eval_accuracy": 0.5955897435897436,
1820
+ "eval_loss": 2.1953744888305664,
1821
+ "eval_runtime": 4.4179,
1822
+ "eval_samples_per_second": 113.177,
1823
+ "eval_steps_per_second": 14.26,
1824
+ "step": 20631
1825
+ },
1826
+ {
1827
+ "epoch": 39.0,
1828
+ "eval_exact_match": 0.0,
1829
+ "eval_f1": 0.0,
1830
+ "step": 20631
1831
+ },
1832
+ {
1833
+ "epoch": 39.13,
1834
+ "learning_rate": 3e-05,
1835
+ "loss": 0.5139,
1836
+ "step": 20700
1837
+ },
1838
+ {
1839
+ "epoch": 39.32,
1840
+ "learning_rate": 3e-05,
1841
+ "loss": 0.5267,
1842
+ "step": 20800
1843
+ },
1844
+ {
1845
+ "epoch": 39.51,
1846
+ "learning_rate": 3e-05,
1847
+ "loss": 0.5189,
1848
+ "step": 20900
1849
+ },
1850
+ {
1851
+ "epoch": 39.7,
1852
+ "learning_rate": 3e-05,
1853
+ "loss": 0.5279,
1854
+ "step": 21000
1855
+ },
1856
+ {
1857
+ "epoch": 39.89,
1858
+ "learning_rate": 3e-05,
1859
+ "loss": 0.5404,
1860
+ "step": 21100
1861
+ },
1862
+ {
1863
+ "epoch": 40.0,
1864
+ "eval_accuracy": 0.5941538461538461,
1865
+ "eval_loss": 2.17238712310791,
1866
+ "eval_runtime": 4.4172,
1867
+ "eval_samples_per_second": 113.194,
1868
+ "eval_steps_per_second": 14.262,
1869
+ "step": 21160
1870
+ },
1871
+ {
1872
+ "epoch": 40.0,
1873
+ "eval_exact_match": 0.0,
1874
+ "eval_f1": 0.0,
1875
+ "step": 21160
1876
+ }
1877
+ ],
1878
+ "logging_steps": 100,
1879
+ "max_steps": 26450,
1880
+ "num_train_epochs": 50,
1881
+ "save_steps": 500,
1882
+ "total_flos": 3.573371429233623e+18,
1883
+ "trial_name": null,
1884
+ "trial_params": null
1885
+ }
checkpoint-21160/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec9dd2a1e083127818a65cb4d99f38a0a803e54a476491027d2e996644eba47c
3
+ size 4728