Hugging Face's logo

linpang
/
alpaca-bitcoin-tweets-sentiment

PEFT
Model card Files Community
2
linpang commited on
Commit
fcb2e5c
•
1 Parent(s): a8b14ca

Upload 27 files

Browse files
Files changed (27) hide show
  1. README.md +3 -0
  2. adapter_config.json +20 -0
  3. adapter_model.bin +3 -0
  4. checkpoint-200/README.md +3 -0
  5. checkpoint-200/adapter_config.json +20 -0
  6. checkpoint-200/adapter_model.bin +3 -0
  7. checkpoint-200/optimizer.pt +3 -0
  8. checkpoint-200/rng_state.pth +3 -0
  9. checkpoint-200/scheduler.pt +3 -0
  10. checkpoint-200/trainer_state.json +168 -0
  11. checkpoint-200/training_args.bin +3 -0
  12. checkpoint-250/README.md +3 -0
  13. checkpoint-250/adapter_config.json +20 -0
  14. checkpoint-250/adapter_model.bin +3 -0
  15. checkpoint-250/optimizer.pt +3 -0
  16. checkpoint-250/rng_state.pth +3 -0
  17. checkpoint-250/scheduler.pt +3 -0
  18. checkpoint-250/trainer_state.json +206 -0
  19. checkpoint-250/training_args.bin +3 -0
  20. checkpoint-300/README.md +3 -0
  21. checkpoint-300/adapter_config.json +20 -0
  22. checkpoint-300/adapter_model.bin +3 -0
  23. checkpoint-300/optimizer.pt +3 -0
  24. checkpoint-300/rng_state.pth +3 -0
  25. checkpoint-300/scheduler.pt +3 -0
  26. checkpoint-300/trainer_state.json +244 -0
  27. checkpoint-300/training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "decapoda-research/llama-7b-hf",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 16,
10
+ "lora_dropout": 0.05,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5e1621f48d9ad8feb1d6d31050275f0aafd080c5c07153301fe2f48411f4406
3
+ size 443
checkpoint-200/README.md ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
checkpoint-200/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "decapoda-research/llama-7b-hf",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 16,
10
+ "lora_dropout": 0.05,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
checkpoint-200/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5e1621f48d9ad8feb1d6d31050275f0aafd080c5c07153301fe2f48411f4406
3
+ size 443
checkpoint-200/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5dcc143e7f40b5b68cbf50adf9df57472edcc18923a5e169cba667149f2f3313
3
+ size 33661637
checkpoint-200/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fbd20306d3fae2c4bf609a9b70a33e0c51c24d79e6e43d91f85a46e1ded4f76c
3
+ size 14575
checkpoint-200/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:733b26a211b260ea1fd3d3e991e0540ad00853a1824d26da983ccafb09e98e4f
3
+ size 627
checkpoint-200/trainer_state.json ADDED
@@ -0,0 +1,168 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6357369422912598,
3
+ "best_model_checkpoint": "experiments/checkpoint-200",
4
+ "epoch": 15.058823529411764,
5
+ "global_step": 200,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.75,
12
+ "learning_rate": 2.9999999999999997e-05,
13
+ "loss": 3.2279,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 1.51,
18
+ "learning_rate": 5.9999999999999995e-05,
19
+ "loss": 3.0585,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 2.26,
24
+ "learning_rate": 8.699999999999999e-05,
25
+ "loss": 2.6599,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 3.01,
30
+ "learning_rate": 0.00011399999999999999,
31
+ "loss": 1.9787,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 3.76,
36
+ "learning_rate": 0.00014099999999999998,
37
+ "loss": 1.378,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 3.76,
42
+ "eval_loss": 1.1715341806411743,
43
+ "eval_runtime": 12.5908,
44
+ "eval_samples_per_second": 15.885,
45
+ "eval_steps_per_second": 1.986,
46
+ "step": 50
47
+ },
48
+ {
49
+ "epoch": 4.52,
50
+ "learning_rate": 0.00017099999999999998,
51
+ "loss": 1.1071,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 5.27,
56
+ "learning_rate": 0.000201,
57
+ "loss": 0.9966,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 6.02,
62
+ "learning_rate": 0.00023099999999999998,
63
+ "loss": 0.9398,
64
+ "step": 80
65
+ },
66
+ {
67
+ "epoch": 6.78,
68
+ "learning_rate": 0.000261,
69
+ "loss": 0.8755,
70
+ "step": 90
71
+ },
72
+ {
73
+ "epoch": 7.53,
74
+ "learning_rate": 0.00029099999999999997,
75
+ "loss": 0.8274,
76
+ "step": 100
77
+ },
78
+ {
79
+ "epoch": 7.53,
80
+ "eval_loss": 0.7494703531265259,
81
+ "eval_runtime": 12.5869,
82
+ "eval_samples_per_second": 15.89,
83
+ "eval_steps_per_second": 1.986,
84
+ "step": 100
85
+ },
86
+ {
87
+ "epoch": 8.28,
88
+ "learning_rate": 0.0002895,
89
+ "loss": 0.761,
90
+ "step": 110
91
+ },
92
+ {
93
+ "epoch": 9.04,
94
+ "learning_rate": 0.0002745,
95
+ "loss": 0.7358,
96
+ "step": 120
97
+ },
98
+ {
99
+ "epoch": 9.79,
100
+ "learning_rate": 0.00025949999999999997,
101
+ "loss": 0.7155,
102
+ "step": 130
103
+ },
104
+ {
105
+ "epoch": 10.54,
106
+ "learning_rate": 0.0002445,
107
+ "loss": 0.6811,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 11.29,
112
+ "learning_rate": 0.0002295,
113
+ "loss": 0.6712,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 11.29,
118
+ "eval_loss": 0.6565232872962952,
119
+ "eval_runtime": 12.5854,
120
+ "eval_samples_per_second": 15.891,
121
+ "eval_steps_per_second": 1.986,
122
+ "step": 150
123
+ },
124
+ {
125
+ "epoch": 12.05,
126
+ "learning_rate": 0.00021449999999999998,
127
+ "loss": 0.6562,
128
+ "step": 160
129
+ },
130
+ {
131
+ "epoch": 12.8,
132
+ "learning_rate": 0.0001995,
133
+ "loss": 0.6367,
134
+ "step": 170
135
+ },
136
+ {
137
+ "epoch": 13.55,
138
+ "learning_rate": 0.00018449999999999999,
139
+ "loss": 0.6173,
140
+ "step": 180
141
+ },
142
+ {
143
+ "epoch": 14.31,
144
+ "learning_rate": 0.00016949999999999997,
145
+ "loss": 0.6267,
146
+ "step": 190
147
+ },
148
+ {
149
+ "epoch": 15.06,
150
+ "learning_rate": 0.0001545,
151
+ "loss": 0.6039,
152
+ "step": 200
153
+ },
154
+ {
155
+ "epoch": 15.06,
156
+ "eval_loss": 0.6357369422912598,
157
+ "eval_runtime": 12.6199,
158
+ "eval_samples_per_second": 15.848,
159
+ "eval_steps_per_second": 1.981,
160
+ "step": 200
161
+ }
162
+ ],
163
+ "max_steps": 300,
164
+ "num_train_epochs": 24,
165
+ "total_flos": 1.3818886480429056e+17,
166
+ "trial_name": null,
167
+ "trial_params": null
168
+ }
checkpoint-200/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b96b63abf71bdba700582ab3fb1568adebe1f8155c04779c47ae46dabab07d8
3
+ size 3899
checkpoint-250/README.md ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
checkpoint-250/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "decapoda-research/llama-7b-hf",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 16,
10
+ "lora_dropout": 0.05,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
checkpoint-250/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5e1621f48d9ad8feb1d6d31050275f0aafd080c5c07153301fe2f48411f4406
3
+ size 443
checkpoint-250/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b0253893fbc9fecc074572e788e858775c30dc40716364411c42e6105293c63b
3
+ size 33661637
checkpoint-250/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bcf89619cc0a3048396a66ae1234c13f6e66934e972b8b52b06bf42d1c8c15f
3
+ size 14575
checkpoint-250/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a93cc0be32e01f149c701aaba527e3d707dabc0197a8fb87648c36a36633978
3
+ size 627
checkpoint-250/trainer_state.json ADDED
@@ -0,0 +1,206 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6293959021568298,
3
+ "best_model_checkpoint": "experiments/checkpoint-250",
4
+ "epoch": 18.823529411764707,
5
+ "global_step": 250,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.75,
12
+ "learning_rate": 2.9999999999999997e-05,
13
+ "loss": 3.2279,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 1.51,
18
+ "learning_rate": 5.9999999999999995e-05,
19
+ "loss": 3.0585,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 2.26,
24
+ "learning_rate": 8.699999999999999e-05,
25
+ "loss": 2.6599,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 3.01,
30
+ "learning_rate": 0.00011399999999999999,
31
+ "loss": 1.9787,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 3.76,
36
+ "learning_rate": 0.00014099999999999998,
37
+ "loss": 1.378,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 3.76,
42
+ "eval_loss": 1.1715341806411743,
43
+ "eval_runtime": 12.5908,
44
+ "eval_samples_per_second": 15.885,
45
+ "eval_steps_per_second": 1.986,
46
+ "step": 50
47
+ },
48
+ {
49
+ "epoch": 4.52,
50
+ "learning_rate": 0.00017099999999999998,
51
+ "loss": 1.1071,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 5.27,
56
+ "learning_rate": 0.000201,
57
+ "loss": 0.9966,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 6.02,
62
+ "learning_rate": 0.00023099999999999998,
63
+ "loss": 0.9398,
64
+ "step": 80
65
+ },
66
+ {
67
+ "epoch": 6.78,
68
+ "learning_rate": 0.000261,
69
+ "loss": 0.8755,
70
+ "step": 90
71
+ },
72
+ {
73
+ "epoch": 7.53,
74
+ "learning_rate": 0.00029099999999999997,
75
+ "loss": 0.8274,
76
+ "step": 100
77
+ },
78
+ {
79
+ "epoch": 7.53,
80
+ "eval_loss": 0.7494703531265259,
81
+ "eval_runtime": 12.5869,
82
+ "eval_samples_per_second": 15.89,
83
+ "eval_steps_per_second": 1.986,
84
+ "step": 100
85
+ },
86
+ {
87
+ "epoch": 8.28,
88
+ "learning_rate": 0.0002895,
89
+ "loss": 0.761,
90
+ "step": 110
91
+ },
92
+ {
93
+ "epoch": 9.04,
94
+ "learning_rate": 0.0002745,
95
+ "loss": 0.7358,
96
+ "step": 120
97
+ },
98
+ {
99
+ "epoch": 9.79,
100
+ "learning_rate": 0.00025949999999999997,
101
+ "loss": 0.7155,
102
+ "step": 130
103
+ },
104
+ {
105
+ "epoch": 10.54,
106
+ "learning_rate": 0.0002445,
107
+ "loss": 0.6811,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 11.29,
112
+ "learning_rate": 0.0002295,
113
+ "loss": 0.6712,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 11.29,
118
+ "eval_loss": 0.6565232872962952,
119
+ "eval_runtime": 12.5854,
120
+ "eval_samples_per_second": 15.891,
121
+ "eval_steps_per_second": 1.986,
122
+ "step": 150
123
+ },
124
+ {
125
+ "epoch": 12.05,
126
+ "learning_rate": 0.00021449999999999998,
127
+ "loss": 0.6562,
128
+ "step": 160
129
+ },
130
+ {
131
+ "epoch": 12.8,
132
+ "learning_rate": 0.0001995,
133
+ "loss": 0.6367,
134
+ "step": 170
135
+ },
136
+ {
137
+ "epoch": 13.55,
138
+ "learning_rate": 0.00018449999999999999,
139
+ "loss": 0.6173,
140
+ "step": 180
141
+ },
142
+ {
143
+ "epoch": 14.31,
144
+ "learning_rate": 0.00016949999999999997,
145
+ "loss": 0.6267,
146
+ "step": 190
147
+ },
148
+ {
149
+ "epoch": 15.06,
150
+ "learning_rate": 0.0001545,
151
+ "loss": 0.6039,
152
+ "step": 200
153
+ },
154
+ {
155
+ "epoch": 15.06,
156
+ "eval_loss": 0.6357369422912598,
157
+ "eval_runtime": 12.6199,
158
+ "eval_samples_per_second": 15.848,
159
+ "eval_steps_per_second": 1.981,
160
+ "step": 200
161
+ },
162
+ {
163
+ "epoch": 15.81,
164
+ "learning_rate": 0.0001395,
165
+ "loss": 0.5949,
166
+ "step": 210
167
+ },
168
+ {
169
+ "epoch": 16.56,
170
+ "learning_rate": 0.0001245,
171
+ "loss": 0.5957,
172
+ "step": 220
173
+ },
174
+ {
175
+ "epoch": 17.32,
176
+ "learning_rate": 0.00010949999999999999,
177
+ "loss": 0.5625,
178
+ "step": 230
179
+ },
180
+ {
181
+ "epoch": 18.07,
182
+ "learning_rate": 9.449999999999999e-05,
183
+ "loss": 0.5897,
184
+ "step": 240
185
+ },
186
+ {
187
+ "epoch": 18.82,
188
+ "learning_rate": 7.95e-05,
189
+ "loss": 0.5747,
190
+ "step": 250
191
+ },
192
+ {
193
+ "epoch": 18.82,
194
+ "eval_loss": 0.6293959021568298,
195
+ "eval_runtime": 12.5951,
196
+ "eval_samples_per_second": 15.879,
197
+ "eval_steps_per_second": 1.985,
198
+ "step": 250
199
+ }
200
+ ],
201
+ "max_steps": 300,
202
+ "num_train_epochs": 24,
203
+ "total_flos": 1.7270585305399296e+17,
204
+ "trial_name": null,
205
+ "trial_params": null
206
+ }
checkpoint-250/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b96b63abf71bdba700582ab3fb1568adebe1f8155c04779c47ae46dabab07d8
3
+ size 3899
checkpoint-300/README.md ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
checkpoint-300/adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "decapoda-research/llama-7b-hf",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "layers_pattern": null,
8
+ "layers_to_transform": null,
9
+ "lora_alpha": 16,
10
+ "lora_dropout": 0.05,
11
+ "modules_to_save": null,
12
+ "peft_type": "LORA",
13
+ "r": 8,
14
+ "revision": null,
15
+ "target_modules": [
16
+ "q_proj",
17
+ "v_proj"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
checkpoint-300/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5e1621f48d9ad8feb1d6d31050275f0aafd080c5c07153301fe2f48411f4406
3
+ size 443
checkpoint-300/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb3c96409d761b403896826e4ab52bfa005158f4d6eb5e3f1db9ea7371919b28
3
+ size 33661637
checkpoint-300/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77f149ad0f17b9e0be901949cde2bcc2f11ae0af037f64aad3814ae9272d5d95
3
+ size 14575
checkpoint-300/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91282cdc194ccfea92191c2ef14e9f21cca2b690f6095e9ad447fd4cded1beec
3
+ size 627
checkpoint-300/trainer_state.json ADDED
@@ -0,0 +1,244 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6293959021568298,
3
+ "best_model_checkpoint": "experiments/checkpoint-250",
4
+ "epoch": 22.58823529411765,
5
+ "global_step": 300,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.75,
12
+ "learning_rate": 2.9999999999999997e-05,
13
+ "loss": 3.2279,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 1.51,
18
+ "learning_rate": 5.9999999999999995e-05,
19
+ "loss": 3.0585,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 2.26,
24
+ "learning_rate": 8.699999999999999e-05,
25
+ "loss": 2.6599,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 3.01,
30
+ "learning_rate": 0.00011399999999999999,
31
+ "loss": 1.9787,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 3.76,
36
+ "learning_rate": 0.00014099999999999998,
37
+ "loss": 1.378,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 3.76,
42
+ "eval_loss": 1.1715341806411743,
43
+ "eval_runtime": 12.5908,
44
+ "eval_samples_per_second": 15.885,
45
+ "eval_steps_per_second": 1.986,
46
+ "step": 50
47
+ },
48
+ {
49
+ "epoch": 4.52,
50
+ "learning_rate": 0.00017099999999999998,
51
+ "loss": 1.1071,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 5.27,
56
+ "learning_rate": 0.000201,
57
+ "loss": 0.9966,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 6.02,
62
+ "learning_rate": 0.00023099999999999998,
63
+ "loss": 0.9398,
64
+ "step": 80
65
+ },
66
+ {
67
+ "epoch": 6.78,
68
+ "learning_rate": 0.000261,
69
+ "loss": 0.8755,
70
+ "step": 90
71
+ },
72
+ {
73
+ "epoch": 7.53,
74
+ "learning_rate": 0.00029099999999999997,
75
+ "loss": 0.8274,
76
+ "step": 100
77
+ },
78
+ {
79
+ "epoch": 7.53,
80
+ "eval_loss": 0.7494703531265259,
81
+ "eval_runtime": 12.5869,
82
+ "eval_samples_per_second": 15.89,
83
+ "eval_steps_per_second": 1.986,
84
+ "step": 100
85
+ },
86
+ {
87
+ "epoch": 8.28,
88
+ "learning_rate": 0.0002895,
89
+ "loss": 0.761,
90
+ "step": 110
91
+ },
92
+ {
93
+ "epoch": 9.04,
94
+ "learning_rate": 0.0002745,
95
+ "loss": 0.7358,
96
+ "step": 120
97
+ },
98
+ {
99
+ "epoch": 9.79,
100
+ "learning_rate": 0.00025949999999999997,
101
+ "loss": 0.7155,
102
+ "step": 130
103
+ },
104
+ {
105
+ "epoch": 10.54,
106
+ "learning_rate": 0.0002445,
107
+ "loss": 0.6811,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 11.29,
112
+ "learning_rate": 0.0002295,
113
+ "loss": 0.6712,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 11.29,
118
+ "eval_loss": 0.6565232872962952,
119
+ "eval_runtime": 12.5854,
120
+ "eval_samples_per_second": 15.891,
121
+ "eval_steps_per_second": 1.986,
122
+ "step": 150
123
+ },
124
+ {
125
+ "epoch": 12.05,
126
+ "learning_rate": 0.00021449999999999998,
127
+ "loss": 0.6562,
128
+ "step": 160
129
+ },
130
+ {
131
+ "epoch": 12.8,
132
+ "learning_rate": 0.0001995,
133
+ "loss": 0.6367,
134
+ "step": 170
135
+ },
136
+ {
137
+ "epoch": 13.55,
138
+ "learning_rate": 0.00018449999999999999,
139
+ "loss": 0.6173,
140
+ "step": 180
141
+ },
142
+ {
143
+ "epoch": 14.31,
144
+ "learning_rate": 0.00016949999999999997,
145
+ "loss": 0.6267,
146
+ "step": 190
147
+ },
148
+ {
149
+ "epoch": 15.06,
150
+ "learning_rate": 0.0001545,
151
+ "loss": 0.6039,
152
+ "step": 200
153
+ },
154
+ {
155
+ "epoch": 15.06,
156
+ "eval_loss": 0.6357369422912598,
157
+ "eval_runtime": 12.6199,
158
+ "eval_samples_per_second": 15.848,
159
+ "eval_steps_per_second": 1.981,
160
+ "step": 200
161
+ },
162
+ {
163
+ "epoch": 15.81,
164
+ "learning_rate": 0.0001395,
165
+ "loss": 0.5949,
166
+ "step": 210
167
+ },
168
+ {
169
+ "epoch": 16.56,
170
+ "learning_rate": 0.0001245,
171
+ "loss": 0.5957,
172
+ "step": 220
173
+ },
174
+ {
175
+ "epoch": 17.32,
176
+ "learning_rate": 0.00010949999999999999,
177
+ "loss": 0.5625,
178
+ "step": 230
179
+ },
180
+ {
181
+ "epoch": 18.07,
182
+ "learning_rate": 9.449999999999999e-05,
183
+ "loss": 0.5897,
184
+ "step": 240
185
+ },
186
+ {
187
+ "epoch": 18.82,
188
+ "learning_rate": 7.95e-05,
189
+ "loss": 0.5747,
190
+ "step": 250
191
+ },
192
+ {
193
+ "epoch": 18.82,
194
+ "eval_loss": 0.6293959021568298,
195
+ "eval_runtime": 12.5951,
196
+ "eval_samples_per_second": 15.879,
197
+ "eval_steps_per_second": 1.985,
198
+ "step": 250
199
+ },
200
+ {
201
+ "epoch": 19.58,
202
+ "learning_rate": 6.45e-05,
203
+ "loss": 0.5588,
204
+ "step": 260
205
+ },
206
+ {
207
+ "epoch": 20.33,
208
+ "learning_rate": 4.95e-05,
209
+ "loss": 0.5501,
210
+ "step": 270
211
+ },
212
+ {
213
+ "epoch": 21.08,
214
+ "learning_rate": 3.45e-05,
215
+ "loss": 0.552,
216
+ "step": 280
217
+ },
218
+ {
219
+ "epoch": 21.84,
220
+ "learning_rate": 1.95e-05,
221
+ "loss": 0.5523,
222
+ "step": 290
223
+ },
224
+ {
225
+ "epoch": 22.59,
226
+ "learning_rate": 4.499999999999999e-06,
227
+ "loss": 0.5447,
228
+ "step": 300
229
+ },
230
+ {
231
+ "epoch": 22.59,
232
+ "eval_loss": 0.633611798286438,
233
+ "eval_runtime": 12.6,
234
+ "eval_samples_per_second": 15.873,
235
+ "eval_steps_per_second": 1.984,
236
+ "step": 300
237
+ }
238
+ ],
239
+ "max_steps": 300,
240
+ "num_train_epochs": 24,
241
+ "total_flos": 2.0721681158111232e+17,
242
+ "trial_name": null,
243
+ "trial_params": null
244
+ }
checkpoint-300/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b96b63abf71bdba700582ab3fb1568adebe1f8155c04779c47ae46dabab07d8
3
+ size 3899