Model save
Browse files- README.md +53 -53
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +2362 -0
- training_args.bin +1 -1
README.md
CHANGED
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
17 |
|
18 |
This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on an unknown dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
-
- Loss: 6.7918
|
21 |
- Accuracy: 0.1930
|
|
|
22 |
|
23 |
## Model description
|
24 |
|
@@ -53,58 +53,58 @@ The following hyperparameters were used during training:
|
|
53 |
|
54 |
### Training results
|
55 |
|
56 |
-
| Training Loss | Epoch | Step | Validation Loss |
|
57 |
-
|
58 |
-
| 1.3891 | 1.0 | 529 | 1.3015 |
|
59 |
-
| 1.3633 | 2.0 | 1058 | 1.2855 |
|
60 |
-
| 1.2929 | 3.0 | 1587 | 1.2954 |
|
61 |
-
| 1.2361 | 4.0 | 2116 | 1.3489 |
|
62 |
-
| 1.1856 | 5.0 | 2645 | 1.3968 |
|
63 |
-
| 1.1098 | 6.0 | 3174 | 1.4721 |
|
64 |
-
| 1.0753 | 7.0 | 3703 | 1.5798 |
|
65 |
-
| 1.0048 | 8.0 | 4232 | 1.6042 |
|
66 |
-
| 0.9456 | 9.0 | 4761 | 1.6843 |
|
67 |
-
| 0.8766 | 10.0 | 5290 | 1.7829 |
|
68 |
-
| 0.8273 | 11.0 | 5819 | 1.8060 |
|
69 |
-
| 0.7755 | 12.0 | 6348 | 1.8729 |
|
70 |
-
| 0.715 | 13.0 | 6877 | 1.9620 |
|
71 |
-
| 0.6804 | 14.0 | 7406 | 2.0030 |
|
72 |
-
| 0.6277 | 15.0 | 7935 | 2.0528 |
|
73 |
-
| 0.5733 | 16.0 | 8464 | 2.0475 |
|
74 |
-
| 0.5409 | 17.0 | 8993 | 2.0920 |
|
75 |
-
| 0.5024 | 18.0 | 9522 | 2.1207 |
|
76 |
-
| 0.4699 | 19.0 | 10051 | 2.1108 |
|
77 |
-
| 0.4367 | 20.0 | 10580 | 2.1089 |
|
78 |
-
| 0.857 | 21.0 | 11109 | 2.0215 |
|
79 |
-
| 3.7434 | 22.0 | 11638 | 10.1186 |
|
80 |
-
| 7.7259 | 23.0 | 12167 | 7.5379 |
|
81 |
-
| 4.2204 | 24.0 | 12696 | 2.1568 |
|
82 |
-
| 0.7385 | 25.0 | 13225 | 1.8229 |
|
83 |
-
| 1.1473 | 26.0 | 13754 | 1.7570 |
|
84 |
-
| 2.0182 | 27.0 | 14283 | 1.7293 |
|
85 |
-
| 2.2707 | 28.0 | 14812 | 2.7017 |
|
86 |
-
| 4.1792 | 29.0 | 15341 | 5.8288 |
|
87 |
-
| 7.7703 | 30.0 | 15870 | 7.6619 |
|
88 |
-
| 7.7034 | 31.0 | 16399 | 7.7003 |
|
89 |
-
| 7.9533 | 32.0 | 16928 | 7.8667 |
|
90 |
-
| 7.8634 | 33.0 | 17457 | 7.8134 |
|
91 |
-
| 7.8584 | 34.0 | 17986 | 7.6760 |
|
92 |
-
| 7.824 | 35.0 | 18515 | 7.7083 |
|
93 |
-
| 7.7446 | 36.0 | 19044 | 7.6626 |
|
94 |
-
| 7.6708 | 37.0 | 19573 | 7.5529 |
|
95 |
-
| 7.6733 | 38.0 | 20102 | 7.5704 |
|
96 |
-
| 7.6271 | 39.0 | 20631 | 7.5363 |
|
97 |
-
| 7.5886 | 40.0 | 21160 | 7.4684 |
|
98 |
-
| 7.5514 | 41.0 | 21689 | 7.4223 |
|
99 |
-
| 7.5205 | 42.0 | 22218 | 7.3508 |
|
100 |
-
| 7.4577 | 43.0 | 22747 | 7.1785 |
|
101 |
-
| 7.5021 | 44.0 | 23276 | 6.6226 |
|
102 |
-
| 7.1133 | 45.0 | 23805 | 6.4100 |
|
103 |
-
| 6.9682 | 46.0 | 24334 | 6.3553 |
|
104 |
-
| 6.8891 | 47.0 | 24863 | 6.3128 |
|
105 |
-
| 6.8621 | 48.0 | 25392 | 6.2465 |
|
106 |
-
| 6.8176 | 49.0 | 25921 | 6.1966 |
|
107 |
-
| 6.9473 | 50.0 | 26450 | 6.7918 |
|
108 |
|
109 |
|
110 |
### Framework versions
|
|
|
17 |
|
18 |
This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on an unknown dataset.
|
19 |
It achieves the following results on the evaluation set:
|
|
|
20 |
- Accuracy: 0.1930
|
21 |
+
- Loss: 6.7918
|
22 |
|
23 |
## Model description
|
24 |
|
|
|
53 |
|
54 |
### Training results
|
55 |
|
56 |
+
| Training Loss | Epoch | Step | Accuracy | Validation Loss |
|
57 |
+
|:-------------:|:-----:|:-----:|:--------:|:---------------:|
|
58 |
+
| 1.3891 | 1.0 | 529 | 0.6138 | 1.3015 |
|
59 |
+
| 1.3633 | 2.0 | 1058 | 0.6166 | 1.2855 |
|
60 |
+
| 1.2929 | 3.0 | 1587 | 0.6177 | 1.2954 |
|
61 |
+
| 1.2361 | 4.0 | 2116 | 0.6045 | 1.3489 |
|
62 |
+
| 1.1856 | 5.0 | 2645 | 0.6125 | 1.3968 |
|
63 |
+
| 1.1098 | 6.0 | 3174 | 0.6115 | 1.4721 |
|
64 |
+
| 1.0753 | 7.0 | 3703 | 0.6076 | 1.5798 |
|
65 |
+
| 1.0048 | 8.0 | 4232 | 0.6084 | 1.6042 |
|
66 |
+
| 0.9456 | 9.0 | 4761 | 0.5977 | 1.6843 |
|
67 |
+
| 0.8766 | 10.0 | 5290 | 0.6051 | 1.7829 |
|
68 |
+
| 0.8273 | 11.0 | 5819 | 0.6043 | 1.8060 |
|
69 |
+
| 0.7755 | 12.0 | 6348 | 0.6019 | 1.8729 |
|
70 |
+
| 0.715 | 13.0 | 6877 | 0.6017 | 1.9620 |
|
71 |
+
| 0.6804 | 14.0 | 7406 | 0.6009 | 2.0030 |
|
72 |
+
| 0.6277 | 15.0 | 7935 | 0.5998 | 2.0528 |
|
73 |
+
| 0.5733 | 16.0 | 8464 | 0.6012 | 2.0475 |
|
74 |
+
| 0.5409 | 17.0 | 8993 | 0.5749 | 2.0920 |
|
75 |
+
| 0.5024 | 18.0 | 9522 | 0.5986 | 2.1207 |
|
76 |
+
| 0.4699 | 19.0 | 10051 | 0.5993 | 2.1108 |
|
77 |
+
| 0.4367 | 20.0 | 10580 | 0.6005 | 2.1089 |
|
78 |
+
| 0.857 | 21.0 | 11109 | 0.5983 | 2.0215 |
|
79 |
+
| 3.7434 | 22.0 | 11638 | 0.2233 | 10.1186 |
|
80 |
+
| 7.7259 | 23.0 | 12167 | 0.1986 | 7.5379 |
|
81 |
+
| 4.2204 | 24.0 | 12696 | 0.5345 | 2.1568 |
|
82 |
+
| 0.7385 | 25.0 | 13225 | 0.5963 | 1.8229 |
|
83 |
+
| 1.1473 | 26.0 | 13754 | 0.5788 | 1.7570 |
|
84 |
+
| 2.0182 | 27.0 | 14283 | 0.5573 | 1.7293 |
|
85 |
+
| 2.2707 | 28.0 | 14812 | 0.4956 | 2.7017 |
|
86 |
+
| 4.1792 | 29.0 | 15341 | 0.3070 | 5.8288 |
|
87 |
+
| 7.7703 | 30.0 | 15870 | 0.1922 | 7.6619 |
|
88 |
+
| 7.7034 | 31.0 | 16399 | 0.1913 | 7.7003 |
|
89 |
+
| 7.9533 | 32.0 | 16928 | 0.1899 | 7.8667 |
|
90 |
+
| 7.8634 | 33.0 | 17457 | 0.1897 | 7.8134 |
|
91 |
+
| 7.8584 | 34.0 | 17986 | 0.1882 | 7.6760 |
|
92 |
+
| 7.824 | 35.0 | 18515 | 0.1888 | 7.7083 |
|
93 |
+
| 7.7446 | 36.0 | 19044 | 0.1888 | 7.6626 |
|
94 |
+
| 7.6708 | 37.0 | 19573 | 0.1886 | 7.5529 |
|
95 |
+
| 7.6733 | 38.0 | 20102 | 0.1903 | 7.5704 |
|
96 |
+
| 7.6271 | 39.0 | 20631 | 0.1949 | 7.5363 |
|
97 |
+
| 7.5886 | 40.0 | 21160 | 0.2130 | 7.4684 |
|
98 |
+
| 7.5514 | 41.0 | 21689 | 0.2077 | 7.4223 |
|
99 |
+
| 7.5205 | 42.0 | 22218 | 0.1946 | 7.3508 |
|
100 |
+
| 7.4577 | 43.0 | 22747 | 0.1951 | 7.1785 |
|
101 |
+
| 7.5021 | 44.0 | 23276 | 0.2092 | 6.6226 |
|
102 |
+
| 7.1133 | 45.0 | 23805 | 0.1994 | 6.4100 |
|
103 |
+
| 6.9682 | 46.0 | 24334 | 0.2250 | 6.3553 |
|
104 |
+
| 6.8891 | 47.0 | 24863 | 0.2224 | 6.3128 |
|
105 |
+
| 6.8621 | 48.0 | 25392 | 0.23 | 6.2465 |
|
106 |
+
| 6.8176 | 49.0 | 25921 | 0.2561 | 6.1966 |
|
107 |
+
| 6.9473 | 50.0 | 26450 | 0.1930 | 6.7918 |
|
108 |
|
109 |
|
110 |
### Framework versions
|
all_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 50.0,
|
3 |
+
"train_loss": 4.017823467146472,
|
4 |
+
"train_runtime": 26175.2328,
|
5 |
+
"train_samples": 16925,
|
6 |
+
"train_samples_per_second": 32.33,
|
7 |
+
"train_steps_per_second": 1.01
|
8 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 50.0,
|
3 |
+
"train_loss": 4.017823467146472,
|
4 |
+
"train_runtime": 26175.2328,
|
5 |
+
"train_samples": 16925,
|
6 |
+
"train_samples_per_second": 32.33,
|
7 |
+
"train_steps_per_second": 1.01
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,2362 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 50.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 26450,
|
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": 5e-05,
|
14 |
+
"loss": 1.6355,
|
15 |
+
"step": 100
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.38,
|
19 |
+
"learning_rate": 5e-05,
|
20 |
+
"loss": 1.4037,
|
21 |
+
"step": 200
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.57,
|
25 |
+
"learning_rate": 5e-05,
|
26 |
+
"loss": 1.4183,
|
27 |
+
"step": 300
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.76,
|
31 |
+
"learning_rate": 5e-05,
|
32 |
+
"loss": 1.3881,
|
33 |
+
"step": 400
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.95,
|
37 |
+
"learning_rate": 5e-05,
|
38 |
+
"loss": 1.3891,
|
39 |
+
"step": 500
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 1.0,
|
43 |
+
"eval_accuracy": 0.6137948717948718,
|
44 |
+
"eval_loss": 1.3015451431274414,
|
45 |
+
"eval_runtime": 4.1821,
|
46 |
+
"eval_samples_per_second": 119.557,
|
47 |
+
"eval_steps_per_second": 15.064,
|
48 |
+
"step": 529
|
49 |
+
},
|
50 |
+
{
|
51 |
+
"epoch": 1.0,
|
52 |
+
"eval_exact_match": 23.2,
|
53 |
+
"eval_f1": 35.17014759120022,
|
54 |
+
"step": 529
|
55 |
+
},
|
56 |
+
{
|
57 |
+
"epoch": 1.13,
|
58 |
+
"learning_rate": 5e-05,
|
59 |
+
"loss": 1.3808,
|
60 |
+
"step": 600
|
61 |
+
},
|
62 |
+
{
|
63 |
+
"epoch": 1.32,
|
64 |
+
"learning_rate": 5e-05,
|
65 |
+
"loss": 1.348,
|
66 |
+
"step": 700
|
67 |
+
},
|
68 |
+
{
|
69 |
+
"epoch": 1.51,
|
70 |
+
"learning_rate": 5e-05,
|
71 |
+
"loss": 1.3627,
|
72 |
+
"step": 800
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 1.7,
|
76 |
+
"learning_rate": 5e-05,
|
77 |
+
"loss": 1.3473,
|
78 |
+
"step": 900
|
79 |
+
},
|
80 |
+
{
|
81 |
+
"epoch": 1.89,
|
82 |
+
"learning_rate": 5e-05,
|
83 |
+
"loss": 1.3633,
|
84 |
+
"step": 1000
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 2.0,
|
88 |
+
"eval_accuracy": 0.6165641025641025,
|
89 |
+
"eval_loss": 1.2854759693145752,
|
90 |
+
"eval_runtime": 4.1064,
|
91 |
+
"eval_samples_per_second": 121.761,
|
92 |
+
"eval_steps_per_second": 15.342,
|
93 |
+
"step": 1058
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 2.0,
|
97 |
+
"eval_exact_match": 25.8,
|
98 |
+
"eval_f1": 37.509047619047635,
|
99 |
+
"step": 1058
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 2.08,
|
103 |
+
"learning_rate": 5e-05,
|
104 |
+
"loss": 1.3358,
|
105 |
+
"step": 1100
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 2.27,
|
109 |
+
"learning_rate": 5e-05,
|
110 |
+
"loss": 1.2917,
|
111 |
+
"step": 1200
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 2.46,
|
115 |
+
"learning_rate": 5e-05,
|
116 |
+
"loss": 1.2948,
|
117 |
+
"step": 1300
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 2.65,
|
121 |
+
"learning_rate": 5e-05,
|
122 |
+
"loss": 1.3144,
|
123 |
+
"step": 1400
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 2.84,
|
127 |
+
"learning_rate": 5e-05,
|
128 |
+
"loss": 1.2929,
|
129 |
+
"step": 1500
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 3.0,
|
133 |
+
"eval_accuracy": 0.6176923076923077,
|
134 |
+
"eval_loss": 1.2953603267669678,
|
135 |
+
"eval_runtime": 4.0992,
|
136 |
+
"eval_samples_per_second": 121.974,
|
137 |
+
"eval_steps_per_second": 15.369,
|
138 |
+
"step": 1587
|
139 |
+
},
|
140 |
+
{
|
141 |
+
"epoch": 3.0,
|
142 |
+
"eval_exact_match": 25.0,
|
143 |
+
"eval_f1": 37.08253968253969,
|
144 |
+
"step": 1587
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 3.02,
|
148 |
+
"learning_rate": 5e-05,
|
149 |
+
"loss": 1.2906,
|
150 |
+
"step": 1600
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 3.21,
|
154 |
+
"learning_rate": 5e-05,
|
155 |
+
"loss": 1.2216,
|
156 |
+
"step": 1700
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 3.4,
|
160 |
+
"learning_rate": 5e-05,
|
161 |
+
"loss": 1.2459,
|
162 |
+
"step": 1800
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 3.59,
|
166 |
+
"learning_rate": 5e-05,
|
167 |
+
"loss": 1.2474,
|
168 |
+
"step": 1900
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 3.78,
|
172 |
+
"learning_rate": 5e-05,
|
173 |
+
"loss": 1.2518,
|
174 |
+
"step": 2000
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 3.97,
|
178 |
+
"learning_rate": 5e-05,
|
179 |
+
"loss": 1.2361,
|
180 |
+
"step": 2100
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 4.0,
|
184 |
+
"eval_accuracy": 0.6045128205128205,
|
185 |
+
"eval_loss": 1.3488503694534302,
|
186 |
+
"eval_runtime": 4.2015,
|
187 |
+
"eval_samples_per_second": 119.004,
|
188 |
+
"eval_steps_per_second": 14.995,
|
189 |
+
"step": 2116
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 4.0,
|
193 |
+
"eval_exact_match": 25.0,
|
194 |
+
"eval_f1": 36.82126984126986,
|
195 |
+
"step": 2116
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 4.16,
|
199 |
+
"learning_rate": 5e-05,
|
200 |
+
"loss": 1.1856,
|
201 |
+
"step": 2200
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 4.35,
|
205 |
+
"learning_rate": 5e-05,
|
206 |
+
"loss": 1.1556,
|
207 |
+
"step": 2300
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 4.54,
|
211 |
+
"learning_rate": 5e-05,
|
212 |
+
"loss": 1.1828,
|
213 |
+
"step": 2400
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 4.73,
|
217 |
+
"learning_rate": 5e-05,
|
218 |
+
"loss": 1.1687,
|
219 |
+
"step": 2500
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 4.91,
|
223 |
+
"learning_rate": 5e-05,
|
224 |
+
"loss": 1.1856,
|
225 |
+
"step": 2600
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 5.0,
|
229 |
+
"eval_accuracy": 0.6125128205128205,
|
230 |
+
"eval_loss": 1.3968186378479004,
|
231 |
+
"eval_runtime": 4.3081,
|
232 |
+
"eval_samples_per_second": 116.06,
|
233 |
+
"eval_steps_per_second": 14.624,
|
234 |
+
"step": 2645
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 5.0,
|
238 |
+
"eval_exact_match": 24.6,
|
239 |
+
"eval_f1": 37.079523809523806,
|
240 |
+
"step": 2645
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 5.1,
|
244 |
+
"learning_rate": 5e-05,
|
245 |
+
"loss": 1.1376,
|
246 |
+
"step": 2700
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 5.29,
|
250 |
+
"learning_rate": 5e-05,
|
251 |
+
"loss": 1.1079,
|
252 |
+
"step": 2800
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 5.48,
|
256 |
+
"learning_rate": 5e-05,
|
257 |
+
"loss": 1.0954,
|
258 |
+
"step": 2900
|
259 |
+
},
|
260 |
+
{
|
261 |
+
"epoch": 5.67,
|
262 |
+
"learning_rate": 5e-05,
|
263 |
+
"loss": 1.1317,
|
264 |
+
"step": 3000
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 5.86,
|
268 |
+
"learning_rate": 5e-05,
|
269 |
+
"loss": 1.1098,
|
270 |
+
"step": 3100
|
271 |
+
},
|
272 |
+
{
|
273 |
+
"epoch": 6.0,
|
274 |
+
"eval_accuracy": 0.6114871794871795,
|
275 |
+
"eval_loss": 1.4720605611801147,
|
276 |
+
"eval_runtime": 4.5,
|
277 |
+
"eval_samples_per_second": 111.11,
|
278 |
+
"eval_steps_per_second": 14.0,
|
279 |
+
"step": 3174
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 6.0,
|
283 |
+
"eval_exact_match": 24.4,
|
284 |
+
"eval_f1": 35.78015873015873,
|
285 |
+
"step": 3174
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 6.05,
|
289 |
+
"learning_rate": 5e-05,
|
290 |
+
"loss": 1.0945,
|
291 |
+
"step": 3200
|
292 |
+
},
|
293 |
+
{
|
294 |
+
"epoch": 6.24,
|
295 |
+
"learning_rate": 5e-05,
|
296 |
+
"loss": 1.0309,
|
297 |
+
"step": 3300
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 6.43,
|
301 |
+
"learning_rate": 5e-05,
|
302 |
+
"loss": 1.0389,
|
303 |
+
"step": 3400
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 6.62,
|
307 |
+
"learning_rate": 5e-05,
|
308 |
+
"loss": 1.0547,
|
309 |
+
"step": 3500
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 6.81,
|
313 |
+
"learning_rate": 5e-05,
|
314 |
+
"loss": 1.0618,
|
315 |
+
"step": 3600
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 6.99,
|
319 |
+
"learning_rate": 5e-05,
|
320 |
+
"loss": 1.0753,
|
321 |
+
"step": 3700
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 7.0,
|
325 |
+
"eval_accuracy": 0.6076410256410256,
|
326 |
+
"eval_loss": 1.5797967910766602,
|
327 |
+
"eval_runtime": 4.8128,
|
328 |
+
"eval_samples_per_second": 103.891,
|
329 |
+
"eval_steps_per_second": 13.09,
|
330 |
+
"step": 3703
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 7.0,
|
334 |
+
"eval_exact_match": 21.8,
|
335 |
+
"eval_f1": 33.15111111111111,
|
336 |
+
"step": 3703
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 7.18,
|
340 |
+
"learning_rate": 5e-05,
|
341 |
+
"loss": 0.9727,
|
342 |
+
"step": 3800
|
343 |
+
},
|
344 |
+
{
|
345 |
+
"epoch": 7.37,
|
346 |
+
"learning_rate": 5e-05,
|
347 |
+
"loss": 0.9815,
|
348 |
+
"step": 3900
|
349 |
+
},
|
350 |
+
{
|
351 |
+
"epoch": 7.56,
|
352 |
+
"learning_rate": 5e-05,
|
353 |
+
"loss": 0.9777,
|
354 |
+
"step": 4000
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 7.75,
|
358 |
+
"learning_rate": 5e-05,
|
359 |
+
"loss": 0.9984,
|
360 |
+
"step": 4100
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 7.94,
|
364 |
+
"learning_rate": 5e-05,
|
365 |
+
"loss": 1.0048,
|
366 |
+
"step": 4200
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 8.0,
|
370 |
+
"eval_accuracy": 0.6084102564102564,
|
371 |
+
"eval_loss": 1.6041829586029053,
|
372 |
+
"eval_runtime": 4.4087,
|
373 |
+
"eval_samples_per_second": 113.411,
|
374 |
+
"eval_steps_per_second": 14.29,
|
375 |
+
"step": 4232
|
376 |
+
},
|
377 |
+
{
|
378 |
+
"epoch": 8.0,
|
379 |
+
"eval_exact_match": 22.4,
|
380 |
+
"eval_f1": 35.09904761904763,
|
381 |
+
"step": 4232
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 8.13,
|
385 |
+
"learning_rate": 5e-05,
|
386 |
+
"loss": 0.9568,
|
387 |
+
"step": 4300
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 8.32,
|
391 |
+
"learning_rate": 5e-05,
|
392 |
+
"loss": 0.9088,
|
393 |
+
"step": 4400
|
394 |
+
},
|
395 |
+
{
|
396 |
+
"epoch": 8.51,
|
397 |
+
"learning_rate": 5e-05,
|
398 |
+
"loss": 0.9341,
|
399 |
+
"step": 4500
|
400 |
+
},
|
401 |
+
{
|
402 |
+
"epoch": 8.7,
|
403 |
+
"learning_rate": 5e-05,
|
404 |
+
"loss": 0.9285,
|
405 |
+
"step": 4600
|
406 |
+
},
|
407 |
+
{
|
408 |
+
"epoch": 8.88,
|
409 |
+
"learning_rate": 5e-05,
|
410 |
+
"loss": 0.9456,
|
411 |
+
"step": 4700
|
412 |
+
},
|
413 |
+
{
|
414 |
+
"epoch": 9.0,
|
415 |
+
"eval_accuracy": 0.5976923076923077,
|
416 |
+
"eval_loss": 1.684310793876648,
|
417 |
+
"eval_runtime": 4.9249,
|
418 |
+
"eval_samples_per_second": 101.524,
|
419 |
+
"eval_steps_per_second": 12.792,
|
420 |
+
"step": 4761
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 9.0,
|
424 |
+
"eval_exact_match": 24.6,
|
425 |
+
"eval_f1": 36.32493506493508,
|
426 |
+
"step": 4761
|
427 |
+
},
|
428 |
+
{
|
429 |
+
"epoch": 9.07,
|
430 |
+
"learning_rate": 5e-05,
|
431 |
+
"loss": 0.912,
|
432 |
+
"step": 4800
|
433 |
+
},
|
434 |
+
{
|
435 |
+
"epoch": 9.26,
|
436 |
+
"learning_rate": 5e-05,
|
437 |
+
"loss": 0.8443,
|
438 |
+
"step": 4900
|
439 |
+
},
|
440 |
+
{
|
441 |
+
"epoch": 9.45,
|
442 |
+
"learning_rate": 5e-05,
|
443 |
+
"loss": 0.8789,
|
444 |
+
"step": 5000
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 9.64,
|
448 |
+
"learning_rate": 5e-05,
|
449 |
+
"loss": 0.8865,
|
450 |
+
"step": 5100
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 9.83,
|
454 |
+
"learning_rate": 5e-05,
|
455 |
+
"loss": 0.8766,
|
456 |
+
"step": 5200
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 10.0,
|
460 |
+
"eval_accuracy": 0.6051282051282051,
|
461 |
+
"eval_loss": 1.7828805446624756,
|
462 |
+
"eval_runtime": 4.1983,
|
463 |
+
"eval_samples_per_second": 119.096,
|
464 |
+
"eval_steps_per_second": 15.006,
|
465 |
+
"step": 5290
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 10.0,
|
469 |
+
"eval_exact_match": 23.0,
|
470 |
+
"eval_f1": 34.336190476190495,
|
471 |
+
"step": 5290
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 10.02,
|
475 |
+
"learning_rate": 5e-05,
|
476 |
+
"loss": 0.8835,
|
477 |
+
"step": 5300
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 10.21,
|
481 |
+
"learning_rate": 5e-05,
|
482 |
+
"loss": 0.8095,
|
483 |
+
"step": 5400
|
484 |
+
},
|
485 |
+
{
|
486 |
+
"epoch": 10.4,
|
487 |
+
"learning_rate": 5e-05,
|
488 |
+
"loss": 0.8008,
|
489 |
+
"step": 5500
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 10.59,
|
493 |
+
"learning_rate": 5e-05,
|
494 |
+
"loss": 0.8208,
|
495 |
+
"step": 5600
|
496 |
+
},
|
497 |
+
{
|
498 |
+
"epoch": 10.78,
|
499 |
+
"learning_rate": 5e-05,
|
500 |
+
"loss": 0.8282,
|
501 |
+
"step": 5700
|
502 |
+
},
|
503 |
+
{
|
504 |
+
"epoch": 10.96,
|
505 |
+
"learning_rate": 5e-05,
|
506 |
+
"loss": 0.8273,
|
507 |
+
"step": 5800
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 11.0,
|
511 |
+
"eval_accuracy": 0.6043076923076923,
|
512 |
+
"eval_loss": 1.8059720993041992,
|
513 |
+
"eval_runtime": 4.2264,
|
514 |
+
"eval_samples_per_second": 118.304,
|
515 |
+
"eval_steps_per_second": 14.906,
|
516 |
+
"step": 5819
|
517 |
+
},
|
518 |
+
{
|
519 |
+
"epoch": 11.0,
|
520 |
+
"eval_exact_match": 21.6,
|
521 |
+
"eval_f1": 32.628095238095234,
|
522 |
+
"step": 5819
|
523 |
+
},
|
524 |
+
{
|
525 |
+
"epoch": 11.15,
|
526 |
+
"learning_rate": 5e-05,
|
527 |
+
"loss": 0.7637,
|
528 |
+
"step": 5900
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 11.34,
|
532 |
+
"learning_rate": 5e-05,
|
533 |
+
"loss": 0.7581,
|
534 |
+
"step": 6000
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 11.53,
|
538 |
+
"learning_rate": 5e-05,
|
539 |
+
"loss": 0.7571,
|
540 |
+
"step": 6100
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 11.72,
|
544 |
+
"learning_rate": 5e-05,
|
545 |
+
"loss": 0.7637,
|
546 |
+
"step": 6200
|
547 |
+
},
|
548 |
+
{
|
549 |
+
"epoch": 11.91,
|
550 |
+
"learning_rate": 5e-05,
|
551 |
+
"loss": 0.7755,
|
552 |
+
"step": 6300
|
553 |
+
},
|
554 |
+
{
|
555 |
+
"epoch": 12.0,
|
556 |
+
"eval_accuracy": 0.6018974358974359,
|
557 |
+
"eval_loss": 1.8729300498962402,
|
558 |
+
"eval_runtime": 4.2001,
|
559 |
+
"eval_samples_per_second": 119.045,
|
560 |
+
"eval_steps_per_second": 15.0,
|
561 |
+
"step": 6348
|
562 |
+
},
|
563 |
+
{
|
564 |
+
"epoch": 12.0,
|
565 |
+
"eval_exact_match": 16.8,
|
566 |
+
"eval_f1": 27.317351339157547,
|
567 |
+
"step": 6348
|
568 |
+
},
|
569 |
+
{
|
570 |
+
"epoch": 12.1,
|
571 |
+
"learning_rate": 5e-05,
|
572 |
+
"loss": 0.7386,
|
573 |
+
"step": 6400
|
574 |
+
},
|
575 |
+
{
|
576 |
+
"epoch": 12.29,
|
577 |
+
"learning_rate": 5e-05,
|
578 |
+
"loss": 0.6977,
|
579 |
+
"step": 6500
|
580 |
+
},
|
581 |
+
{
|
582 |
+
"epoch": 12.48,
|
583 |
+
"learning_rate": 5e-05,
|
584 |
+
"loss": 0.7185,
|
585 |
+
"step": 6600
|
586 |
+
},
|
587 |
+
{
|
588 |
+
"epoch": 12.67,
|
589 |
+
"learning_rate": 5e-05,
|
590 |
+
"loss": 0.7166,
|
591 |
+
"step": 6700
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 12.85,
|
595 |
+
"learning_rate": 5e-05,
|
596 |
+
"loss": 0.715,
|
597 |
+
"step": 6800
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 13.0,
|
601 |
+
"eval_accuracy": 0.6016923076923076,
|
602 |
+
"eval_loss": 1.9620097875595093,
|
603 |
+
"eval_runtime": 4.4055,
|
604 |
+
"eval_samples_per_second": 113.495,
|
605 |
+
"eval_steps_per_second": 14.3,
|
606 |
+
"step": 6877
|
607 |
+
},
|
608 |
+
{
|
609 |
+
"epoch": 13.0,
|
610 |
+
"eval_exact_match": 4.8,
|
611 |
+
"eval_f1": 9.356456790950029,
|
612 |
+
"step": 6877
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 13.04,
|
616 |
+
"learning_rate": 5e-05,
|
617 |
+
"loss": 0.7032,
|
618 |
+
"step": 6900
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 13.23,
|
622 |
+
"learning_rate": 5e-05,
|
623 |
+
"loss": 0.6494,
|
624 |
+
"step": 7000
|
625 |
+
},
|
626 |
+
{
|
627 |
+
"epoch": 13.42,
|
628 |
+
"learning_rate": 5e-05,
|
629 |
+
"loss": 0.6584,
|
630 |
+
"step": 7100
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 13.61,
|
634 |
+
"learning_rate": 5e-05,
|
635 |
+
"loss": 0.6681,
|
636 |
+
"step": 7200
|
637 |
+
},
|
638 |
+
{
|
639 |
+
"epoch": 13.8,
|
640 |
+
"learning_rate": 5e-05,
|
641 |
+
"loss": 0.6556,
|
642 |
+
"step": 7300
|
643 |
+
},
|
644 |
+
{
|
645 |
+
"epoch": 13.99,
|
646 |
+
"learning_rate": 5e-05,
|
647 |
+
"loss": 0.6804,
|
648 |
+
"step": 7400
|
649 |
+
},
|
650 |
+
{
|
651 |
+
"epoch": 14.0,
|
652 |
+
"eval_accuracy": 0.6009230769230769,
|
653 |
+
"eval_loss": 2.00297474861145,
|
654 |
+
"eval_runtime": 4.2009,
|
655 |
+
"eval_samples_per_second": 119.021,
|
656 |
+
"eval_steps_per_second": 14.997,
|
657 |
+
"step": 7406
|
658 |
+
},
|
659 |
+
{
|
660 |
+
"epoch": 14.0,
|
661 |
+
"eval_exact_match": 5.8,
|
662 |
+
"eval_f1": 13.996400415302853,
|
663 |
+
"step": 7406
|
664 |
+
},
|
665 |
+
{
|
666 |
+
"epoch": 14.18,
|
667 |
+
"learning_rate": 5e-05,
|
668 |
+
"loss": 0.5983,
|
669 |
+
"step": 7500
|
670 |
+
},
|
671 |
+
{
|
672 |
+
"epoch": 14.37,
|
673 |
+
"learning_rate": 5e-05,
|
674 |
+
"loss": 0.6056,
|
675 |
+
"step": 7600
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 14.56,
|
679 |
+
"learning_rate": 5e-05,
|
680 |
+
"loss": 0.6048,
|
681 |
+
"step": 7700
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 14.74,
|
685 |
+
"learning_rate": 5e-05,
|
686 |
+
"loss": 0.6322,
|
687 |
+
"step": 7800
|
688 |
+
},
|
689 |
+
{
|
690 |
+
"epoch": 14.93,
|
691 |
+
"learning_rate": 5e-05,
|
692 |
+
"loss": 0.6277,
|
693 |
+
"step": 7900
|
694 |
+
},
|
695 |
+
{
|
696 |
+
"epoch": 15.0,
|
697 |
+
"eval_accuracy": 0.5998461538461538,
|
698 |
+
"eval_loss": 2.0527751445770264,
|
699 |
+
"eval_runtime": 4.4287,
|
700 |
+
"eval_samples_per_second": 112.9,
|
701 |
+
"eval_steps_per_second": 14.225,
|
702 |
+
"step": 7935
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 15.0,
|
706 |
+
"eval_exact_match": 3.2,
|
707 |
+
"eval_f1": 8.78350649350649,
|
708 |
+
"step": 7935
|
709 |
+
},
|
710 |
+
{
|
711 |
+
"epoch": 15.12,
|
712 |
+
"learning_rate": 5e-05,
|
713 |
+
"loss": 0.5776,
|
714 |
+
"step": 8000
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 15.31,
|
718 |
+
"learning_rate": 5e-05,
|
719 |
+
"loss": 0.5522,
|
720 |
+
"step": 8100
|
721 |
+
},
|
722 |
+
{
|
723 |
+
"epoch": 15.5,
|
724 |
+
"learning_rate": 5e-05,
|
725 |
+
"loss": 0.5735,
|
726 |
+
"step": 8200
|
727 |
+
},
|
728 |
+
{
|
729 |
+
"epoch": 15.69,
|
730 |
+
"learning_rate": 5e-05,
|
731 |
+
"loss": 0.5637,
|
732 |
+
"step": 8300
|
733 |
+
},
|
734 |
+
{
|
735 |
+
"epoch": 15.88,
|
736 |
+
"learning_rate": 5e-05,
|
737 |
+
"loss": 0.5733,
|
738 |
+
"step": 8400
|
739 |
+
},
|
740 |
+
{
|
741 |
+
"epoch": 16.0,
|
742 |
+
"eval_accuracy": 0.6012307692307692,
|
743 |
+
"eval_loss": 2.047515392303467,
|
744 |
+
"eval_runtime": 4.1989,
|
745 |
+
"eval_samples_per_second": 119.079,
|
746 |
+
"eval_steps_per_second": 15.004,
|
747 |
+
"step": 8464
|
748 |
+
},
|
749 |
+
{
|
750 |
+
"epoch": 16.0,
|
751 |
+
"eval_exact_match": 0.6,
|
752 |
+
"eval_f1": 1.077443805758639,
|
753 |
+
"step": 8464
|
754 |
+
},
|
755 |
+
{
|
756 |
+
"epoch": 16.07,
|
757 |
+
"learning_rate": 5e-05,
|
758 |
+
"loss": 0.5601,
|
759 |
+
"step": 8500
|
760 |
+
},
|
761 |
+
{
|
762 |
+
"epoch": 16.26,
|
763 |
+
"learning_rate": 5e-05,
|
764 |
+
"loss": 0.5126,
|
765 |
+
"step": 8600
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 16.45,
|
769 |
+
"learning_rate": 5e-05,
|
770 |
+
"loss": 0.5217,
|
771 |
+
"step": 8700
|
772 |
+
},
|
773 |
+
{
|
774 |
+
"epoch": 16.64,
|
775 |
+
"learning_rate": 5e-05,
|
776 |
+
"loss": 0.5291,
|
777 |
+
"step": 8800
|
778 |
+
},
|
779 |
+
{
|
780 |
+
"epoch": 16.82,
|
781 |
+
"learning_rate": 5e-05,
|
782 |
+
"loss": 0.5409,
|
783 |
+
"step": 8900
|
784 |
+
},
|
785 |
+
{
|
786 |
+
"epoch": 17.0,
|
787 |
+
"eval_accuracy": 0.5748717948717949,
|
788 |
+
"eval_loss": 2.092020273208618,
|
789 |
+
"eval_runtime": 4.1044,
|
790 |
+
"eval_samples_per_second": 121.822,
|
791 |
+
"eval_steps_per_second": 15.35,
|
792 |
+
"step": 8993
|
793 |
+
},
|
794 |
+
{
|
795 |
+
"epoch": 17.0,
|
796 |
+
"eval_exact_match": 0.4,
|
797 |
+
"eval_f1": 2.398489535596529,
|
798 |
+
"step": 8993
|
799 |
+
},
|
800 |
+
{
|
801 |
+
"epoch": 17.01,
|
802 |
+
"learning_rate": 5e-05,
|
803 |
+
"loss": 0.5354,
|
804 |
+
"step": 9000
|
805 |
+
},
|
806 |
+
{
|
807 |
+
"epoch": 17.2,
|
808 |
+
"learning_rate": 5e-05,
|
809 |
+
"loss": 0.4629,
|
810 |
+
"step": 9100
|
811 |
+
},
|
812 |
+
{
|
813 |
+
"epoch": 17.39,
|
814 |
+
"learning_rate": 5e-05,
|
815 |
+
"loss": 0.4795,
|
816 |
+
"step": 9200
|
817 |
+
},
|
818 |
+
{
|
819 |
+
"epoch": 17.58,
|
820 |
+
"learning_rate": 5e-05,
|
821 |
+
"loss": 0.4921,
|
822 |
+
"step": 9300
|
823 |
+
},
|
824 |
+
{
|
825 |
+
"epoch": 17.77,
|
826 |
+
"learning_rate": 5e-05,
|
827 |
+
"loss": 0.4928,
|
828 |
+
"step": 9400
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 17.96,
|
832 |
+
"learning_rate": 5e-05,
|
833 |
+
"loss": 0.5024,
|
834 |
+
"step": 9500
|
835 |
+
},
|
836 |
+
{
|
837 |
+
"epoch": 18.0,
|
838 |
+
"eval_accuracy": 0.5986153846153847,
|
839 |
+
"eval_loss": 2.120666980743408,
|
840 |
+
"eval_runtime": 4.3114,
|
841 |
+
"eval_samples_per_second": 115.973,
|
842 |
+
"eval_steps_per_second": 14.613,
|
843 |
+
"step": 9522
|
844 |
+
},
|
845 |
+
{
|
846 |
+
"epoch": 18.0,
|
847 |
+
"eval_exact_match": 0.2,
|
848 |
+
"eval_f1": 0.6343183616587786,
|
849 |
+
"step": 9522
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 18.15,
|
853 |
+
"learning_rate": 5e-05,
|
854 |
+
"loss": 0.4539,
|
855 |
+
"step": 9600
|
856 |
+
},
|
857 |
+
{
|
858 |
+
"epoch": 18.34,
|
859 |
+
"learning_rate": 5e-05,
|
860 |
+
"loss": 0.4345,
|
861 |
+
"step": 9700
|
862 |
+
},
|
863 |
+
{
|
864 |
+
"epoch": 18.53,
|
865 |
+
"learning_rate": 5e-05,
|
866 |
+
"loss": 0.4399,
|
867 |
+
"step": 9800
|
868 |
+
},
|
869 |
+
{
|
870 |
+
"epoch": 18.71,
|
871 |
+
"learning_rate": 5e-05,
|
872 |
+
"loss": 0.4581,
|
873 |
+
"step": 9900
|
874 |
+
},
|
875 |
+
{
|
876 |
+
"epoch": 18.9,
|
877 |
+
"learning_rate": 5e-05,
|
878 |
+
"loss": 0.4699,
|
879 |
+
"step": 10000
|
880 |
+
},
|
881 |
+
{
|
882 |
+
"epoch": 19.0,
|
883 |
+
"eval_accuracy": 0.5992820512820513,
|
884 |
+
"eval_loss": 2.110786199569702,
|
885 |
+
"eval_runtime": 4.2008,
|
886 |
+
"eval_samples_per_second": 119.024,
|
887 |
+
"eval_steps_per_second": 14.997,
|
888 |
+
"step": 10051
|
889 |
+
},
|
890 |
+
{
|
891 |
+
"epoch": 19.0,
|
892 |
+
"eval_exact_match": 0.0,
|
893 |
+
"eval_f1": 0.006153846153846154,
|
894 |
+
"step": 10051
|
895 |
+
},
|
896 |
+
{
|
897 |
+
"epoch": 19.09,
|
898 |
+
"learning_rate": 5e-05,
|
899 |
+
"loss": 0.4363,
|
900 |
+
"step": 10100
|
901 |
+
},
|
902 |
+
{
|
903 |
+
"epoch": 19.28,
|
904 |
+
"learning_rate": 5e-05,
|
905 |
+
"loss": 0.4021,
|
906 |
+
"step": 10200
|
907 |
+
},
|
908 |
+
{
|
909 |
+
"epoch": 19.47,
|
910 |
+
"learning_rate": 5e-05,
|
911 |
+
"loss": 0.4161,
|
912 |
+
"step": 10300
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 19.66,
|
916 |
+
"learning_rate": 5e-05,
|
917 |
+
"loss": 0.425,
|
918 |
+
"step": 10400
|
919 |
+
},
|
920 |
+
{
|
921 |
+
"epoch": 19.85,
|
922 |
+
"learning_rate": 5e-05,
|
923 |
+
"loss": 0.4367,
|
924 |
+
"step": 10500
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 20.0,
|
928 |
+
"eval_accuracy": 0.6004615384615385,
|
929 |
+
"eval_loss": 2.1088976860046387,
|
930 |
+
"eval_runtime": 4.0958,
|
931 |
+
"eval_samples_per_second": 122.078,
|
932 |
+
"eval_steps_per_second": 15.382,
|
933 |
+
"step": 10580
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 20.0,
|
937 |
+
"eval_exact_match": 0.0,
|
938 |
+
"eval_f1": 0.04062256613300257,
|
939 |
+
"step": 10580
|
940 |
+
},
|
941 |
+
{
|
942 |
+
"epoch": 20.04,
|
943 |
+
"learning_rate": 5e-05,
|
944 |
+
"loss": 0.4264,
|
945 |
+
"step": 10600
|
946 |
+
},
|
947 |
+
{
|
948 |
+
"epoch": 20.23,
|
949 |
+
"learning_rate": 5e-05,
|
950 |
+
"loss": 0.3746,
|
951 |
+
"step": 10700
|
952 |
+
},
|
953 |
+
{
|
954 |
+
"epoch": 20.42,
|
955 |
+
"learning_rate": 5e-05,
|
956 |
+
"loss": 0.3834,
|
957 |
+
"step": 10800
|
958 |
+
},
|
959 |
+
{
|
960 |
+
"epoch": 20.6,
|
961 |
+
"learning_rate": 5e-05,
|
962 |
+
"loss": 0.4015,
|
963 |
+
"step": 10900
|
964 |
+
},
|
965 |
+
{
|
966 |
+
"epoch": 20.79,
|
967 |
+
"learning_rate": 5e-05,
|
968 |
+
"loss": 0.4284,
|
969 |
+
"step": 11000
|
970 |
+
},
|
971 |
+
{
|
972 |
+
"epoch": 20.98,
|
973 |
+
"learning_rate": 5e-05,
|
974 |
+
"loss": 0.857,
|
975 |
+
"step": 11100
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 21.0,
|
979 |
+
"eval_accuracy": 0.5983076923076923,
|
980 |
+
"eval_loss": 2.0215413570404053,
|
981 |
+
"eval_runtime": 4.0942,
|
982 |
+
"eval_samples_per_second": 122.125,
|
983 |
+
"eval_steps_per_second": 15.388,
|
984 |
+
"step": 11109
|
985 |
+
},
|
986 |
+
{
|
987 |
+
"epoch": 21.0,
|
988 |
+
"eval_exact_match": 0.0,
|
989 |
+
"eval_f1": 0.0,
|
990 |
+
"step": 11109
|
991 |
+
},
|
992 |
+
{
|
993 |
+
"epoch": 21.17,
|
994 |
+
"learning_rate": 5e-05,
|
995 |
+
"loss": 0.8884,
|
996 |
+
"step": 11200
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 21.36,
|
1000 |
+
"learning_rate": 5e-05,
|
1001 |
+
"loss": 1.4123,
|
1002 |
+
"step": 11300
|
1003 |
+
},
|
1004 |
+
{
|
1005 |
+
"epoch": 21.55,
|
1006 |
+
"learning_rate": 5e-05,
|
1007 |
+
"loss": 2.3033,
|
1008 |
+
"step": 11400
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 21.74,
|
1012 |
+
"learning_rate": 5e-05,
|
1013 |
+
"loss": 2.1144,
|
1014 |
+
"step": 11500
|
1015 |
+
},
|
1016 |
+
{
|
1017 |
+
"epoch": 21.93,
|
1018 |
+
"learning_rate": 5e-05,
|
1019 |
+
"loss": 3.7434,
|
1020 |
+
"step": 11600
|
1021 |
+
},
|
1022 |
+
{
|
1023 |
+
"epoch": 22.0,
|
1024 |
+
"eval_accuracy": 0.22333333333333333,
|
1025 |
+
"eval_loss": 10.118596076965332,
|
1026 |
+
"eval_runtime": 4.3085,
|
1027 |
+
"eval_samples_per_second": 116.049,
|
1028 |
+
"eval_steps_per_second": 14.622,
|
1029 |
+
"step": 11638
|
1030 |
+
},
|
1031 |
+
{
|
1032 |
+
"epoch": 22.0,
|
1033 |
+
"eval_exact_match": 0.0,
|
1034 |
+
"eval_f1": 0.0,
|
1035 |
+
"step": 11638
|
1036 |
+
},
|
1037 |
+
{
|
1038 |
+
"epoch": 22.12,
|
1039 |
+
"learning_rate": 5e-05,
|
1040 |
+
"loss": 9.1896,
|
1041 |
+
"step": 11700
|
1042 |
+
},
|
1043 |
+
{
|
1044 |
+
"epoch": 22.31,
|
1045 |
+
"learning_rate": 5e-05,
|
1046 |
+
"loss": 9.6027,
|
1047 |
+
"step": 11800
|
1048 |
+
},
|
1049 |
+
{
|
1050 |
+
"epoch": 22.5,
|
1051 |
+
"learning_rate": 5e-05,
|
1052 |
+
"loss": 11.4681,
|
1053 |
+
"step": 11900
|
1054 |
+
},
|
1055 |
+
{
|
1056 |
+
"epoch": 22.68,
|
1057 |
+
"learning_rate": 5e-05,
|
1058 |
+
"loss": 8.4061,
|
1059 |
+
"step": 12000
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 22.87,
|
1063 |
+
"learning_rate": 5e-05,
|
1064 |
+
"loss": 7.7259,
|
1065 |
+
"step": 12100
|
1066 |
+
},
|
1067 |
+
{
|
1068 |
+
"epoch": 23.0,
|
1069 |
+
"eval_accuracy": 0.19856410256410256,
|
1070 |
+
"eval_loss": 7.537852764129639,
|
1071 |
+
"eval_runtime": 4.1062,
|
1072 |
+
"eval_samples_per_second": 121.768,
|
1073 |
+
"eval_steps_per_second": 15.343,
|
1074 |
+
"step": 12167
|
1075 |
+
},
|
1076 |
+
{
|
1077 |
+
"epoch": 23.0,
|
1078 |
+
"eval_exact_match": 0.0,
|
1079 |
+
"eval_f1": 0.08333333333333333,
|
1080 |
+
"step": 12167
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 23.06,
|
1084 |
+
"learning_rate": 5e-05,
|
1085 |
+
"loss": 7.3869,
|
1086 |
+
"step": 12200
|
1087 |
+
},
|
1088 |
+
{
|
1089 |
+
"epoch": 23.25,
|
1090 |
+
"learning_rate": 5e-05,
|
1091 |
+
"loss": 7.3438,
|
1092 |
+
"step": 12300
|
1093 |
+
},
|
1094 |
+
{
|
1095 |
+
"epoch": 23.44,
|
1096 |
+
"learning_rate": 5e-05,
|
1097 |
+
"loss": 7.1672,
|
1098 |
+
"step": 12400
|
1099 |
+
},
|
1100 |
+
{
|
1101 |
+
"epoch": 23.63,
|
1102 |
+
"learning_rate": 5e-05,
|
1103 |
+
"loss": 7.5678,
|
1104 |
+
"step": 12500
|
1105 |
+
},
|
1106 |
+
{
|
1107 |
+
"epoch": 23.82,
|
1108 |
+
"learning_rate": 5e-05,
|
1109 |
+
"loss": 4.2204,
|
1110 |
+
"step": 12600
|
1111 |
+
},
|
1112 |
+
{
|
1113 |
+
"epoch": 24.0,
|
1114 |
+
"eval_accuracy": 0.5344615384615384,
|
1115 |
+
"eval_loss": 2.1567773818969727,
|
1116 |
+
"eval_runtime": 4.8187,
|
1117 |
+
"eval_samples_per_second": 103.763,
|
1118 |
+
"eval_steps_per_second": 13.074,
|
1119 |
+
"step": 12696
|
1120 |
+
},
|
1121 |
+
{
|
1122 |
+
"epoch": 24.0,
|
1123 |
+
"eval_exact_match": 0.0,
|
1124 |
+
"eval_f1": 0.1556396274796546,
|
1125 |
+
"step": 12696
|
1126 |
+
},
|
1127 |
+
{
|
1128 |
+
"epoch": 24.01,
|
1129 |
+
"learning_rate": 5e-05,
|
1130 |
+
"loss": 2.243,
|
1131 |
+
"step": 12700
|
1132 |
+
},
|
1133 |
+
{
|
1134 |
+
"epoch": 24.2,
|
1135 |
+
"learning_rate": 5e-05,
|
1136 |
+
"loss": 1.4067,
|
1137 |
+
"step": 12800
|
1138 |
+
},
|
1139 |
+
{
|
1140 |
+
"epoch": 24.39,
|
1141 |
+
"learning_rate": 5e-05,
|
1142 |
+
"loss": 1.0065,
|
1143 |
+
"step": 12900
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 24.57,
|
1147 |
+
"learning_rate": 5e-05,
|
1148 |
+
"loss": 0.8179,
|
1149 |
+
"step": 13000
|
1150 |
+
},
|
1151 |
+
{
|
1152 |
+
"epoch": 24.76,
|
1153 |
+
"learning_rate": 5e-05,
|
1154 |
+
"loss": 0.7917,
|
1155 |
+
"step": 13100
|
1156 |
+
},
|
1157 |
+
{
|
1158 |
+
"epoch": 24.95,
|
1159 |
+
"learning_rate": 5e-05,
|
1160 |
+
"loss": 0.7385,
|
1161 |
+
"step": 13200
|
1162 |
+
},
|
1163 |
+
{
|
1164 |
+
"epoch": 25.0,
|
1165 |
+
"eval_accuracy": 0.5963076923076923,
|
1166 |
+
"eval_loss": 1.8228886127471924,
|
1167 |
+
"eval_runtime": 4.8179,
|
1168 |
+
"eval_samples_per_second": 103.781,
|
1169 |
+
"eval_steps_per_second": 13.076,
|
1170 |
+
"step": 13225
|
1171 |
+
},
|
1172 |
+
{
|
1173 |
+
"epoch": 25.0,
|
1174 |
+
"eval_exact_match": 5.2,
|
1175 |
+
"eval_f1": 8.718379522680063,
|
1176 |
+
"step": 13225
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 25.14,
|
1180 |
+
"learning_rate": 5e-05,
|
1181 |
+
"loss": 0.8862,
|
1182 |
+
"step": 13300
|
1183 |
+
},
|
1184 |
+
{
|
1185 |
+
"epoch": 25.33,
|
1186 |
+
"learning_rate": 5e-05,
|
1187 |
+
"loss": 0.5963,
|
1188 |
+
"step": 13400
|
1189 |
+
},
|
1190 |
+
{
|
1191 |
+
"epoch": 25.52,
|
1192 |
+
"learning_rate": 5e-05,
|
1193 |
+
"loss": 2.7783,
|
1194 |
+
"step": 13500
|
1195 |
+
},
|
1196 |
+
{
|
1197 |
+
"epoch": 25.71,
|
1198 |
+
"learning_rate": 5e-05,
|
1199 |
+
"loss": 1.4347,
|
1200 |
+
"step": 13600
|
1201 |
+
},
|
1202 |
+
{
|
1203 |
+
"epoch": 25.9,
|
1204 |
+
"learning_rate": 5e-05,
|
1205 |
+
"loss": 1.1473,
|
1206 |
+
"step": 13700
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 26.0,
|
1210 |
+
"eval_accuracy": 0.5788205128205128,
|
1211 |
+
"eval_loss": 1.7569822072982788,
|
1212 |
+
"eval_runtime": 4.1999,
|
1213 |
+
"eval_samples_per_second": 119.049,
|
1214 |
+
"eval_steps_per_second": 15.0,
|
1215 |
+
"step": 13754
|
1216 |
+
},
|
1217 |
+
{
|
1218 |
+
"epoch": 26.0,
|
1219 |
+
"eval_exact_match": 0.0,
|
1220 |
+
"eval_f1": 0.0,
|
1221 |
+
"step": 13754
|
1222 |
+
},
|
1223 |
+
{
|
1224 |
+
"epoch": 26.09,
|
1225 |
+
"learning_rate": 5e-05,
|
1226 |
+
"loss": 1.2078,
|
1227 |
+
"step": 13800
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 26.28,
|
1231 |
+
"learning_rate": 5e-05,
|
1232 |
+
"loss": 1.2602,
|
1233 |
+
"step": 13900
|
1234 |
+
},
|
1235 |
+
{
|
1236 |
+
"epoch": 26.47,
|
1237 |
+
"learning_rate": 5e-05,
|
1238 |
+
"loss": 2.3188,
|
1239 |
+
"step": 14000
|
1240 |
+
},
|
1241 |
+
{
|
1242 |
+
"epoch": 26.65,
|
1243 |
+
"learning_rate": 5e-05,
|
1244 |
+
"loss": 2.2177,
|
1245 |
+
"step": 14100
|
1246 |
+
},
|
1247 |
+
{
|
1248 |
+
"epoch": 26.84,
|
1249 |
+
"learning_rate": 5e-05,
|
1250 |
+
"loss": 2.0182,
|
1251 |
+
"step": 14200
|
1252 |
+
},
|
1253 |
+
{
|
1254 |
+
"epoch": 27.0,
|
1255 |
+
"eval_accuracy": 0.5573333333333333,
|
1256 |
+
"eval_loss": 1.729324221611023,
|
1257 |
+
"eval_runtime": 4.5012,
|
1258 |
+
"eval_samples_per_second": 111.082,
|
1259 |
+
"eval_steps_per_second": 13.996,
|
1260 |
+
"step": 14283
|
1261 |
+
},
|
1262 |
+
{
|
1263 |
+
"epoch": 27.0,
|
1264 |
+
"eval_exact_match": 0.0,
|
1265 |
+
"eval_f1": 0.03798603689725877,
|
1266 |
+
"step": 14283
|
1267 |
+
},
|
1268 |
+
{
|
1269 |
+
"epoch": 27.03,
|
1270 |
+
"learning_rate": 5e-05,
|
1271 |
+
"loss": 1.4892,
|
1272 |
+
"step": 14300
|
1273 |
+
},
|
1274 |
+
{
|
1275 |
+
"epoch": 27.22,
|
1276 |
+
"learning_rate": 5e-05,
|
1277 |
+
"loss": 1.6886,
|
1278 |
+
"step": 14400
|
1279 |
+
},
|
1280 |
+
{
|
1281 |
+
"epoch": 27.41,
|
1282 |
+
"learning_rate": 5e-05,
|
1283 |
+
"loss": 2.2033,
|
1284 |
+
"step": 14500
|
1285 |
+
},
|
1286 |
+
{
|
1287 |
+
"epoch": 27.6,
|
1288 |
+
"learning_rate": 5e-05,
|
1289 |
+
"loss": 2.5533,
|
1290 |
+
"step": 14600
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 27.79,
|
1294 |
+
"learning_rate": 5e-05,
|
1295 |
+
"loss": 2.5524,
|
1296 |
+
"step": 14700
|
1297 |
+
},
|
1298 |
+
{
|
1299 |
+
"epoch": 27.98,
|
1300 |
+
"learning_rate": 5e-05,
|
1301 |
+
"loss": 2.2707,
|
1302 |
+
"step": 14800
|
1303 |
+
},
|
1304 |
+
{
|
1305 |
+
"epoch": 28.0,
|
1306 |
+
"eval_accuracy": 0.4955897435897436,
|
1307 |
+
"eval_loss": 2.7017431259155273,
|
1308 |
+
"eval_runtime": 4.2078,
|
1309 |
+
"eval_samples_per_second": 118.828,
|
1310 |
+
"eval_steps_per_second": 14.972,
|
1311 |
+
"step": 14812
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 28.0,
|
1315 |
+
"eval_exact_match": 0.0,
|
1316 |
+
"eval_f1": 0.0059701492537313425,
|
1317 |
+
"step": 14812
|
1318 |
+
},
|
1319 |
+
{
|
1320 |
+
"epoch": 28.17,
|
1321 |
+
"learning_rate": 5e-05,
|
1322 |
+
"loss": 2.4127,
|
1323 |
+
"step": 14900
|
1324 |
+
},
|
1325 |
+
{
|
1326 |
+
"epoch": 28.36,
|
1327 |
+
"learning_rate": 5e-05,
|
1328 |
+
"loss": 2.3233,
|
1329 |
+
"step": 15000
|
1330 |
+
},
|
1331 |
+
{
|
1332 |
+
"epoch": 28.54,
|
1333 |
+
"learning_rate": 5e-05,
|
1334 |
+
"loss": 2.2381,
|
1335 |
+
"step": 15100
|
1336 |
+
},
|
1337 |
+
{
|
1338 |
+
"epoch": 28.73,
|
1339 |
+
"learning_rate": 5e-05,
|
1340 |
+
"loss": 2.9648,
|
1341 |
+
"step": 15200
|
1342 |
+
},
|
1343 |
+
{
|
1344 |
+
"epoch": 28.92,
|
1345 |
+
"learning_rate": 5e-05,
|
1346 |
+
"loss": 4.1792,
|
1347 |
+
"step": 15300
|
1348 |
+
},
|
1349 |
+
{
|
1350 |
+
"epoch": 29.0,
|
1351 |
+
"eval_accuracy": 0.307025641025641,
|
1352 |
+
"eval_loss": 5.828761577606201,
|
1353 |
+
"eval_runtime": 4.2948,
|
1354 |
+
"eval_samples_per_second": 116.419,
|
1355 |
+
"eval_steps_per_second": 14.669,
|
1356 |
+
"step": 15341
|
1357 |
+
},
|
1358 |
+
{
|
1359 |
+
"epoch": 29.0,
|
1360 |
+
"eval_exact_match": 0.0,
|
1361 |
+
"eval_f1": 0.0,
|
1362 |
+
"step": 15341
|
1363 |
+
},
|
1364 |
+
{
|
1365 |
+
"epoch": 29.11,
|
1366 |
+
"learning_rate": 5e-05,
|
1367 |
+
"loss": 4.8652,
|
1368 |
+
"step": 15400
|
1369 |
+
},
|
1370 |
+
{
|
1371 |
+
"epoch": 29.3,
|
1372 |
+
"learning_rate": 5e-05,
|
1373 |
+
"loss": 5.5675,
|
1374 |
+
"step": 15500
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 29.49,
|
1378 |
+
"learning_rate": 5e-05,
|
1379 |
+
"loss": 6.4337,
|
1380 |
+
"step": 15600
|
1381 |
+
},
|
1382 |
+
{
|
1383 |
+
"epoch": 29.68,
|
1384 |
+
"learning_rate": 5e-05,
|
1385 |
+
"loss": 7.5369,
|
1386 |
+
"step": 15700
|
1387 |
+
},
|
1388 |
+
{
|
1389 |
+
"epoch": 29.87,
|
1390 |
+
"learning_rate": 5e-05,
|
1391 |
+
"loss": 7.7703,
|
1392 |
+
"step": 15800
|
1393 |
+
},
|
1394 |
+
{
|
1395 |
+
"epoch": 30.0,
|
1396 |
+
"eval_accuracy": 0.19215384615384615,
|
1397 |
+
"eval_loss": 7.661909103393555,
|
1398 |
+
"eval_runtime": 4.4106,
|
1399 |
+
"eval_samples_per_second": 113.363,
|
1400 |
+
"eval_steps_per_second": 14.284,
|
1401 |
+
"step": 15870
|
1402 |
+
},
|
1403 |
+
{
|
1404 |
+
"epoch": 30.0,
|
1405 |
+
"eval_exact_match": 0.0,
|
1406 |
+
"eval_f1": 0.0,
|
1407 |
+
"step": 15870
|
1408 |
+
},
|
1409 |
+
{
|
1410 |
+
"epoch": 30.06,
|
1411 |
+
"learning_rate": 5e-05,
|
1412 |
+
"loss": 7.7483,
|
1413 |
+
"step": 15900
|
1414 |
+
},
|
1415 |
+
{
|
1416 |
+
"epoch": 30.25,
|
1417 |
+
"learning_rate": 5e-05,
|
1418 |
+
"loss": 7.6746,
|
1419 |
+
"step": 16000
|
1420 |
+
},
|
1421 |
+
{
|
1422 |
+
"epoch": 30.43,
|
1423 |
+
"learning_rate": 5e-05,
|
1424 |
+
"loss": 7.603,
|
1425 |
+
"step": 16100
|
1426 |
+
},
|
1427 |
+
{
|
1428 |
+
"epoch": 30.62,
|
1429 |
+
"learning_rate": 5e-05,
|
1430 |
+
"loss": 7.7505,
|
1431 |
+
"step": 16200
|
1432 |
+
},
|
1433 |
+
{
|
1434 |
+
"epoch": 30.81,
|
1435 |
+
"learning_rate": 5e-05,
|
1436 |
+
"loss": 7.7034,
|
1437 |
+
"step": 16300
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 31.0,
|
1441 |
+
"eval_accuracy": 0.19128205128205128,
|
1442 |
+
"eval_loss": 7.700345516204834,
|
1443 |
+
"eval_runtime": 4.103,
|
1444 |
+
"eval_samples_per_second": 121.863,
|
1445 |
+
"eval_steps_per_second": 15.355,
|
1446 |
+
"step": 16399
|
1447 |
+
},
|
1448 |
+
{
|
1449 |
+
"epoch": 31.0,
|
1450 |
+
"eval_exact_match": 0.0,
|
1451 |
+
"eval_f1": 0.0,
|
1452 |
+
"step": 16399
|
1453 |
+
},
|
1454 |
+
{
|
1455 |
+
"epoch": 31.0,
|
1456 |
+
"learning_rate": 5e-05,
|
1457 |
+
"loss": 7.6791,
|
1458 |
+
"step": 16400
|
1459 |
+
},
|
1460 |
+
{
|
1461 |
+
"epoch": 31.19,
|
1462 |
+
"learning_rate": 5e-05,
|
1463 |
+
"loss": 7.8604,
|
1464 |
+
"step": 16500
|
1465 |
+
},
|
1466 |
+
{
|
1467 |
+
"epoch": 31.38,
|
1468 |
+
"learning_rate": 5e-05,
|
1469 |
+
"loss": 7.8378,
|
1470 |
+
"step": 16600
|
1471 |
+
},
|
1472 |
+
{
|
1473 |
+
"epoch": 31.57,
|
1474 |
+
"learning_rate": 5e-05,
|
1475 |
+
"loss": 7.893,
|
1476 |
+
"step": 16700
|
1477 |
+
},
|
1478 |
+
{
|
1479 |
+
"epoch": 31.76,
|
1480 |
+
"learning_rate": 5e-05,
|
1481 |
+
"loss": 7.9123,
|
1482 |
+
"step": 16800
|
1483 |
+
},
|
1484 |
+
{
|
1485 |
+
"epoch": 31.95,
|
1486 |
+
"learning_rate": 5e-05,
|
1487 |
+
"loss": 7.9533,
|
1488 |
+
"step": 16900
|
1489 |
+
},
|
1490 |
+
{
|
1491 |
+
"epoch": 32.0,
|
1492 |
+
"eval_accuracy": 0.18994871794871795,
|
1493 |
+
"eval_loss": 7.8667449951171875,
|
1494 |
+
"eval_runtime": 4.0996,
|
1495 |
+
"eval_samples_per_second": 121.963,
|
1496 |
+
"eval_steps_per_second": 15.367,
|
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": 5e-05,
|
1508 |
+
"loss": 7.9955,
|
1509 |
+
"step": 17000
|
1510 |
+
},
|
1511 |
+
{
|
1512 |
+
"epoch": 32.33,
|
1513 |
+
"learning_rate": 5e-05,
|
1514 |
+
"loss": 7.9227,
|
1515 |
+
"step": 17100
|
1516 |
+
},
|
1517 |
+
{
|
1518 |
+
"epoch": 32.51,
|
1519 |
+
"learning_rate": 5e-05,
|
1520 |
+
"loss": 7.9429,
|
1521 |
+
"step": 17200
|
1522 |
+
},
|
1523 |
+
{
|
1524 |
+
"epoch": 32.7,
|
1525 |
+
"learning_rate": 5e-05,
|
1526 |
+
"loss": 7.8826,
|
1527 |
+
"step": 17300
|
1528 |
+
},
|
1529 |
+
{
|
1530 |
+
"epoch": 32.89,
|
1531 |
+
"learning_rate": 5e-05,
|
1532 |
+
"loss": 7.8634,
|
1533 |
+
"step": 17400
|
1534 |
+
},
|
1535 |
+
{
|
1536 |
+
"epoch": 33.0,
|
1537 |
+
"eval_accuracy": 0.18974358974358974,
|
1538 |
+
"eval_loss": 7.813361167907715,
|
1539 |
+
"eval_runtime": 4.3681,
|
1540 |
+
"eval_samples_per_second": 114.465,
|
1541 |
+
"eval_steps_per_second": 14.423,
|
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": 5e-05,
|
1553 |
+
"loss": 7.8948,
|
1554 |
+
"step": 17500
|
1555 |
+
},
|
1556 |
+
{
|
1557 |
+
"epoch": 33.27,
|
1558 |
+
"learning_rate": 5e-05,
|
1559 |
+
"loss": 7.8637,
|
1560 |
+
"step": 17600
|
1561 |
+
},
|
1562 |
+
{
|
1563 |
+
"epoch": 33.46,
|
1564 |
+
"learning_rate": 5e-05,
|
1565 |
+
"loss": 7.9794,
|
1566 |
+
"step": 17700
|
1567 |
+
},
|
1568 |
+
{
|
1569 |
+
"epoch": 33.65,
|
1570 |
+
"learning_rate": 5e-05,
|
1571 |
+
"loss": 7.9124,
|
1572 |
+
"step": 17800
|
1573 |
+
},
|
1574 |
+
{
|
1575 |
+
"epoch": 33.84,
|
1576 |
+
"learning_rate": 5e-05,
|
1577 |
+
"loss": 7.8584,
|
1578 |
+
"step": 17900
|
1579 |
+
},
|
1580 |
+
{
|
1581 |
+
"epoch": 34.0,
|
1582 |
+
"eval_accuracy": 0.1882051282051282,
|
1583 |
+
"eval_loss": 7.675987243652344,
|
1584 |
+
"eval_runtime": 4.8156,
|
1585 |
+
"eval_samples_per_second": 103.83,
|
1586 |
+
"eval_steps_per_second": 13.083,
|
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": 5e-05,
|
1598 |
+
"loss": 7.7803,
|
1599 |
+
"step": 18000
|
1600 |
+
},
|
1601 |
+
{
|
1602 |
+
"epoch": 34.22,
|
1603 |
+
"learning_rate": 5e-05,
|
1604 |
+
"loss": 7.831,
|
1605 |
+
"step": 18100
|
1606 |
+
},
|
1607 |
+
{
|
1608 |
+
"epoch": 34.4,
|
1609 |
+
"learning_rate": 5e-05,
|
1610 |
+
"loss": 7.7742,
|
1611 |
+
"step": 18200
|
1612 |
+
},
|
1613 |
+
{
|
1614 |
+
"epoch": 34.59,
|
1615 |
+
"learning_rate": 5e-05,
|
1616 |
+
"loss": 7.7111,
|
1617 |
+
"step": 18300
|
1618 |
+
},
|
1619 |
+
{
|
1620 |
+
"epoch": 34.78,
|
1621 |
+
"learning_rate": 5e-05,
|
1622 |
+
"loss": 7.7003,
|
1623 |
+
"step": 18400
|
1624 |
+
},
|
1625 |
+
{
|
1626 |
+
"epoch": 34.97,
|
1627 |
+
"learning_rate": 5e-05,
|
1628 |
+
"loss": 7.824,
|
1629 |
+
"step": 18500
|
1630 |
+
},
|
1631 |
+
{
|
1632 |
+
"epoch": 35.0,
|
1633 |
+
"eval_accuracy": 0.18882051282051282,
|
1634 |
+
"eval_loss": 7.708338737487793,
|
1635 |
+
"eval_runtime": 4.4094,
|
1636 |
+
"eval_samples_per_second": 113.394,
|
1637 |
+
"eval_steps_per_second": 14.288,
|
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": 5e-05,
|
1649 |
+
"loss": 7.7841,
|
1650 |
+
"step": 18600
|
1651 |
+
},
|
1652 |
+
{
|
1653 |
+
"epoch": 35.35,
|
1654 |
+
"learning_rate": 5e-05,
|
1655 |
+
"loss": 7.7835,
|
1656 |
+
"step": 18700
|
1657 |
+
},
|
1658 |
+
{
|
1659 |
+
"epoch": 35.54,
|
1660 |
+
"learning_rate": 5e-05,
|
1661 |
+
"loss": 7.7445,
|
1662 |
+
"step": 18800
|
1663 |
+
},
|
1664 |
+
{
|
1665 |
+
"epoch": 35.73,
|
1666 |
+
"learning_rate": 5e-05,
|
1667 |
+
"loss": 7.7534,
|
1668 |
+
"step": 18900
|
1669 |
+
},
|
1670 |
+
{
|
1671 |
+
"epoch": 35.92,
|
1672 |
+
"learning_rate": 5e-05,
|
1673 |
+
"loss": 7.7446,
|
1674 |
+
"step": 19000
|
1675 |
+
},
|
1676 |
+
{
|
1677 |
+
"epoch": 36.0,
|
1678 |
+
"eval_accuracy": 0.18876923076923077,
|
1679 |
+
"eval_loss": 7.662600517272949,
|
1680 |
+
"eval_runtime": 4.1047,
|
1681 |
+
"eval_samples_per_second": 121.813,
|
1682 |
+
"eval_steps_per_second": 15.348,
|
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": 5e-05,
|
1694 |
+
"loss": 7.7429,
|
1695 |
+
"step": 19100
|
1696 |
+
},
|
1697 |
+
{
|
1698 |
+
"epoch": 36.29,
|
1699 |
+
"learning_rate": 5e-05,
|
1700 |
+
"loss": 7.7358,
|
1701 |
+
"step": 19200
|
1702 |
+
},
|
1703 |
+
{
|
1704 |
+
"epoch": 36.48,
|
1705 |
+
"learning_rate": 5e-05,
|
1706 |
+
"loss": 7.7383,
|
1707 |
+
"step": 19300
|
1708 |
+
},
|
1709 |
+
{
|
1710 |
+
"epoch": 36.67,
|
1711 |
+
"learning_rate": 5e-05,
|
1712 |
+
"loss": 7.6699,
|
1713 |
+
"step": 19400
|
1714 |
+
},
|
1715 |
+
{
|
1716 |
+
"epoch": 36.86,
|
1717 |
+
"learning_rate": 5e-05,
|
1718 |
+
"loss": 7.6708,
|
1719 |
+
"step": 19500
|
1720 |
+
},
|
1721 |
+
{
|
1722 |
+
"epoch": 37.0,
|
1723 |
+
"eval_accuracy": 0.18861538461538463,
|
1724 |
+
"eval_loss": 7.552917957305908,
|
1725 |
+
"eval_runtime": 4.2063,
|
1726 |
+
"eval_samples_per_second": 118.871,
|
1727 |
+
"eval_steps_per_second": 14.978,
|
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": 5e-05,
|
1739 |
+
"loss": 7.6666,
|
1740 |
+
"step": 19600
|
1741 |
+
},
|
1742 |
+
{
|
1743 |
+
"epoch": 37.24,
|
1744 |
+
"learning_rate": 5e-05,
|
1745 |
+
"loss": 7.6806,
|
1746 |
+
"step": 19700
|
1747 |
+
},
|
1748 |
+
{
|
1749 |
+
"epoch": 37.43,
|
1750 |
+
"learning_rate": 5e-05,
|
1751 |
+
"loss": 7.6722,
|
1752 |
+
"step": 19800
|
1753 |
+
},
|
1754 |
+
{
|
1755 |
+
"epoch": 37.62,
|
1756 |
+
"learning_rate": 5e-05,
|
1757 |
+
"loss": 7.6849,
|
1758 |
+
"step": 19900
|
1759 |
+
},
|
1760 |
+
{
|
1761 |
+
"epoch": 37.81,
|
1762 |
+
"learning_rate": 5e-05,
|
1763 |
+
"loss": 7.6704,
|
1764 |
+
"step": 20000
|
1765 |
+
},
|
1766 |
+
{
|
1767 |
+
"epoch": 38.0,
|
1768 |
+
"learning_rate": 5e-05,
|
1769 |
+
"loss": 7.6733,
|
1770 |
+
"step": 20100
|
1771 |
+
},
|
1772 |
+
{
|
1773 |
+
"epoch": 38.0,
|
1774 |
+
"eval_accuracy": 0.19030769230769232,
|
1775 |
+
"eval_loss": 7.570416450500488,
|
1776 |
+
"eval_runtime": 4.8393,
|
1777 |
+
"eval_samples_per_second": 103.32,
|
1778 |
+
"eval_steps_per_second": 13.018,
|
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": 5e-05,
|
1790 |
+
"loss": 7.6549,
|
1791 |
+
"step": 20200
|
1792 |
+
},
|
1793 |
+
{
|
1794 |
+
"epoch": 38.37,
|
1795 |
+
"learning_rate": 5e-05,
|
1796 |
+
"loss": 7.6237,
|
1797 |
+
"step": 20300
|
1798 |
+
},
|
1799 |
+
{
|
1800 |
+
"epoch": 38.56,
|
1801 |
+
"learning_rate": 5e-05,
|
1802 |
+
"loss": 7.6291,
|
1803 |
+
"step": 20400
|
1804 |
+
},
|
1805 |
+
{
|
1806 |
+
"epoch": 38.75,
|
1807 |
+
"learning_rate": 5e-05,
|
1808 |
+
"loss": 7.6129,
|
1809 |
+
"step": 20500
|
1810 |
+
},
|
1811 |
+
{
|
1812 |
+
"epoch": 38.94,
|
1813 |
+
"learning_rate": 5e-05,
|
1814 |
+
"loss": 7.6271,
|
1815 |
+
"step": 20600
|
1816 |
+
},
|
1817 |
+
{
|
1818 |
+
"epoch": 39.0,
|
1819 |
+
"eval_accuracy": 0.19487179487179487,
|
1820 |
+
"eval_loss": 7.5362548828125,
|
1821 |
+
"eval_runtime": 4.4106,
|
1822 |
+
"eval_samples_per_second": 113.363,
|
1823 |
+
"eval_steps_per_second": 14.284,
|
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": 5e-05,
|
1835 |
+
"loss": 7.6365,
|
1836 |
+
"step": 20700
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 39.32,
|
1840 |
+
"learning_rate": 5e-05,
|
1841 |
+
"loss": 7.6218,
|
1842 |
+
"step": 20800
|
1843 |
+
},
|
1844 |
+
{
|
1845 |
+
"epoch": 39.51,
|
1846 |
+
"learning_rate": 5e-05,
|
1847 |
+
"loss": 7.6271,
|
1848 |
+
"step": 20900
|
1849 |
+
},
|
1850 |
+
{
|
1851 |
+
"epoch": 39.7,
|
1852 |
+
"learning_rate": 5e-05,
|
1853 |
+
"loss": 7.6249,
|
1854 |
+
"step": 21000
|
1855 |
+
},
|
1856 |
+
{
|
1857 |
+
"epoch": 39.89,
|
1858 |
+
"learning_rate": 5e-05,
|
1859 |
+
"loss": 7.5886,
|
1860 |
+
"step": 21100
|
1861 |
+
},
|
1862 |
+
{
|
1863 |
+
"epoch": 40.0,
|
1864 |
+
"eval_accuracy": 0.21302564102564103,
|
1865 |
+
"eval_loss": 7.468353748321533,
|
1866 |
+
"eval_runtime": 4.5982,
|
1867 |
+
"eval_samples_per_second": 108.738,
|
1868 |
+
"eval_steps_per_second": 13.701,
|
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 |
+
"epoch": 40.08,
|
1879 |
+
"learning_rate": 5e-05,
|
1880 |
+
"loss": 7.5842,
|
1881 |
+
"step": 21200
|
1882 |
+
},
|
1883 |
+
{
|
1884 |
+
"epoch": 40.26,
|
1885 |
+
"learning_rate": 5e-05,
|
1886 |
+
"loss": 7.5987,
|
1887 |
+
"step": 21300
|
1888 |
+
},
|
1889 |
+
{
|
1890 |
+
"epoch": 40.45,
|
1891 |
+
"learning_rate": 5e-05,
|
1892 |
+
"loss": 7.5694,
|
1893 |
+
"step": 21400
|
1894 |
+
},
|
1895 |
+
{
|
1896 |
+
"epoch": 40.64,
|
1897 |
+
"learning_rate": 5e-05,
|
1898 |
+
"loss": 7.5779,
|
1899 |
+
"step": 21500
|
1900 |
+
},
|
1901 |
+
{
|
1902 |
+
"epoch": 40.83,
|
1903 |
+
"learning_rate": 5e-05,
|
1904 |
+
"loss": 7.5514,
|
1905 |
+
"step": 21600
|
1906 |
+
},
|
1907 |
+
{
|
1908 |
+
"epoch": 41.0,
|
1909 |
+
"eval_accuracy": 0.20774358974358975,
|
1910 |
+
"eval_loss": 7.422276496887207,
|
1911 |
+
"eval_runtime": 4.2316,
|
1912 |
+
"eval_samples_per_second": 118.16,
|
1913 |
+
"eval_steps_per_second": 14.888,
|
1914 |
+
"step": 21689
|
1915 |
+
},
|
1916 |
+
{
|
1917 |
+
"epoch": 41.0,
|
1918 |
+
"eval_exact_match": 0.0,
|
1919 |
+
"eval_f1": 0.0,
|
1920 |
+
"step": 21689
|
1921 |
+
},
|
1922 |
+
{
|
1923 |
+
"epoch": 41.02,
|
1924 |
+
"learning_rate": 5e-05,
|
1925 |
+
"loss": 7.5404,
|
1926 |
+
"step": 21700
|
1927 |
+
},
|
1928 |
+
{
|
1929 |
+
"epoch": 41.21,
|
1930 |
+
"learning_rate": 5e-05,
|
1931 |
+
"loss": 7.5435,
|
1932 |
+
"step": 21800
|
1933 |
+
},
|
1934 |
+
{
|
1935 |
+
"epoch": 41.4,
|
1936 |
+
"learning_rate": 5e-05,
|
1937 |
+
"loss": 7.526,
|
1938 |
+
"step": 21900
|
1939 |
+
},
|
1940 |
+
{
|
1941 |
+
"epoch": 41.59,
|
1942 |
+
"learning_rate": 5e-05,
|
1943 |
+
"loss": 7.5238,
|
1944 |
+
"step": 22000
|
1945 |
+
},
|
1946 |
+
{
|
1947 |
+
"epoch": 41.78,
|
1948 |
+
"learning_rate": 5e-05,
|
1949 |
+
"loss": 7.5161,
|
1950 |
+
"step": 22100
|
1951 |
+
},
|
1952 |
+
{
|
1953 |
+
"epoch": 41.97,
|
1954 |
+
"learning_rate": 5e-05,
|
1955 |
+
"loss": 7.5205,
|
1956 |
+
"step": 22200
|
1957 |
+
},
|
1958 |
+
{
|
1959 |
+
"epoch": 42.0,
|
1960 |
+
"eval_accuracy": 0.19456410256410256,
|
1961 |
+
"eval_loss": 7.350826263427734,
|
1962 |
+
"eval_runtime": 5.0373,
|
1963 |
+
"eval_samples_per_second": 99.259,
|
1964 |
+
"eval_steps_per_second": 12.507,
|
1965 |
+
"step": 22218
|
1966 |
+
},
|
1967 |
+
{
|
1968 |
+
"epoch": 42.0,
|
1969 |
+
"eval_exact_match": 0.0,
|
1970 |
+
"eval_f1": 0.0,
|
1971 |
+
"step": 22218
|
1972 |
+
},
|
1973 |
+
{
|
1974 |
+
"epoch": 42.16,
|
1975 |
+
"learning_rate": 5e-05,
|
1976 |
+
"loss": 7.486,
|
1977 |
+
"step": 22300
|
1978 |
+
},
|
1979 |
+
{
|
1980 |
+
"epoch": 42.34,
|
1981 |
+
"learning_rate": 5e-05,
|
1982 |
+
"loss": 7.4922,
|
1983 |
+
"step": 22400
|
1984 |
+
},
|
1985 |
+
{
|
1986 |
+
"epoch": 42.53,
|
1987 |
+
"learning_rate": 5e-05,
|
1988 |
+
"loss": 7.4647,
|
1989 |
+
"step": 22500
|
1990 |
+
},
|
1991 |
+
{
|
1992 |
+
"epoch": 42.72,
|
1993 |
+
"learning_rate": 5e-05,
|
1994 |
+
"loss": 7.4584,
|
1995 |
+
"step": 22600
|
1996 |
+
},
|
1997 |
+
{
|
1998 |
+
"epoch": 42.91,
|
1999 |
+
"learning_rate": 5e-05,
|
2000 |
+
"loss": 7.4577,
|
2001 |
+
"step": 22700
|
2002 |
+
},
|
2003 |
+
{
|
2004 |
+
"epoch": 43.0,
|
2005 |
+
"eval_accuracy": 0.19512820512820514,
|
2006 |
+
"eval_loss": 7.178500175476074,
|
2007 |
+
"eval_runtime": 4.1027,
|
2008 |
+
"eval_samples_per_second": 121.872,
|
2009 |
+
"eval_steps_per_second": 15.356,
|
2010 |
+
"step": 22747
|
2011 |
+
},
|
2012 |
+
{
|
2013 |
+
"epoch": 43.0,
|
2014 |
+
"eval_exact_match": 0.0,
|
2015 |
+
"eval_f1": 0.0,
|
2016 |
+
"step": 22747
|
2017 |
+
},
|
2018 |
+
{
|
2019 |
+
"epoch": 43.1,
|
2020 |
+
"learning_rate": 5e-05,
|
2021 |
+
"loss": 7.4186,
|
2022 |
+
"step": 22800
|
2023 |
+
},
|
2024 |
+
{
|
2025 |
+
"epoch": 43.29,
|
2026 |
+
"learning_rate": 5e-05,
|
2027 |
+
"loss": 7.4057,
|
2028 |
+
"step": 22900
|
2029 |
+
},
|
2030 |
+
{
|
2031 |
+
"epoch": 43.48,
|
2032 |
+
"learning_rate": 5e-05,
|
2033 |
+
"loss": 7.4082,
|
2034 |
+
"step": 23000
|
2035 |
+
},
|
2036 |
+
{
|
2037 |
+
"epoch": 43.67,
|
2038 |
+
"learning_rate": 5e-05,
|
2039 |
+
"loss": 7.3923,
|
2040 |
+
"step": 23100
|
2041 |
+
},
|
2042 |
+
{
|
2043 |
+
"epoch": 43.86,
|
2044 |
+
"learning_rate": 5e-05,
|
2045 |
+
"loss": 7.5021,
|
2046 |
+
"step": 23200
|
2047 |
+
},
|
2048 |
+
{
|
2049 |
+
"epoch": 44.0,
|
2050 |
+
"eval_accuracy": 0.20923076923076922,
|
2051 |
+
"eval_loss": 6.622580051422119,
|
2052 |
+
"eval_runtime": 5.127,
|
2053 |
+
"eval_samples_per_second": 97.522,
|
2054 |
+
"eval_steps_per_second": 12.288,
|
2055 |
+
"step": 23276
|
2056 |
+
},
|
2057 |
+
{
|
2058 |
+
"epoch": 44.0,
|
2059 |
+
"eval_exact_match": 0.0,
|
2060 |
+
"eval_f1": 0.0,
|
2061 |
+
"step": 23276
|
2062 |
+
},
|
2063 |
+
{
|
2064 |
+
"epoch": 44.05,
|
2065 |
+
"learning_rate": 5e-05,
|
2066 |
+
"loss": 7.0886,
|
2067 |
+
"step": 23300
|
2068 |
+
},
|
2069 |
+
{
|
2070 |
+
"epoch": 44.23,
|
2071 |
+
"learning_rate": 5e-05,
|
2072 |
+
"loss": 6.8981,
|
2073 |
+
"step": 23400
|
2074 |
+
},
|
2075 |
+
{
|
2076 |
+
"epoch": 44.42,
|
2077 |
+
"learning_rate": 5e-05,
|
2078 |
+
"loss": 6.935,
|
2079 |
+
"step": 23500
|
2080 |
+
},
|
2081 |
+
{
|
2082 |
+
"epoch": 44.61,
|
2083 |
+
"learning_rate": 5e-05,
|
2084 |
+
"loss": 7.391,
|
2085 |
+
"step": 23600
|
2086 |
+
},
|
2087 |
+
{
|
2088 |
+
"epoch": 44.8,
|
2089 |
+
"learning_rate": 5e-05,
|
2090 |
+
"loss": 7.9261,
|
2091 |
+
"step": 23700
|
2092 |
+
},
|
2093 |
+
{
|
2094 |
+
"epoch": 44.99,
|
2095 |
+
"learning_rate": 5e-05,
|
2096 |
+
"loss": 7.1133,
|
2097 |
+
"step": 23800
|
2098 |
+
},
|
2099 |
+
{
|
2100 |
+
"epoch": 45.0,
|
2101 |
+
"eval_accuracy": 0.19943589743589743,
|
2102 |
+
"eval_loss": 6.4100470542907715,
|
2103 |
+
"eval_runtime": 4.1044,
|
2104 |
+
"eval_samples_per_second": 121.82,
|
2105 |
+
"eval_steps_per_second": 15.349,
|
2106 |
+
"step": 23805
|
2107 |
+
},
|
2108 |
+
{
|
2109 |
+
"epoch": 45.0,
|
2110 |
+
"eval_exact_match": 0.0,
|
2111 |
+
"eval_f1": 0.0,
|
2112 |
+
"step": 23805
|
2113 |
+
},
|
2114 |
+
{
|
2115 |
+
"epoch": 45.18,
|
2116 |
+
"learning_rate": 5e-05,
|
2117 |
+
"loss": 7.0339,
|
2118 |
+
"step": 23900
|
2119 |
+
},
|
2120 |
+
{
|
2121 |
+
"epoch": 45.37,
|
2122 |
+
"learning_rate": 5e-05,
|
2123 |
+
"loss": 7.4855,
|
2124 |
+
"step": 24000
|
2125 |
+
},
|
2126 |
+
{
|
2127 |
+
"epoch": 45.56,
|
2128 |
+
"learning_rate": 5e-05,
|
2129 |
+
"loss": 7.0461,
|
2130 |
+
"step": 24100
|
2131 |
+
},
|
2132 |
+
{
|
2133 |
+
"epoch": 45.75,
|
2134 |
+
"learning_rate": 5e-05,
|
2135 |
+
"loss": 7.0083,
|
2136 |
+
"step": 24200
|
2137 |
+
},
|
2138 |
+
{
|
2139 |
+
"epoch": 45.94,
|
2140 |
+
"learning_rate": 5e-05,
|
2141 |
+
"loss": 6.9682,
|
2142 |
+
"step": 24300
|
2143 |
+
},
|
2144 |
+
{
|
2145 |
+
"epoch": 46.0,
|
2146 |
+
"eval_accuracy": 0.22497435897435897,
|
2147 |
+
"eval_loss": 6.355289459228516,
|
2148 |
+
"eval_runtime": 4.1015,
|
2149 |
+
"eval_samples_per_second": 121.905,
|
2150 |
+
"eval_steps_per_second": 15.36,
|
2151 |
+
"step": 24334
|
2152 |
+
},
|
2153 |
+
{
|
2154 |
+
"epoch": 46.0,
|
2155 |
+
"eval_exact_match": 0.0,
|
2156 |
+
"eval_f1": 0.0,
|
2157 |
+
"step": 24334
|
2158 |
+
},
|
2159 |
+
{
|
2160 |
+
"epoch": 46.12,
|
2161 |
+
"learning_rate": 5e-05,
|
2162 |
+
"loss": 6.9544,
|
2163 |
+
"step": 24400
|
2164 |
+
},
|
2165 |
+
{
|
2166 |
+
"epoch": 46.31,
|
2167 |
+
"learning_rate": 5e-05,
|
2168 |
+
"loss": 6.9313,
|
2169 |
+
"step": 24500
|
2170 |
+
},
|
2171 |
+
{
|
2172 |
+
"epoch": 46.5,
|
2173 |
+
"learning_rate": 5e-05,
|
2174 |
+
"loss": 6.9282,
|
2175 |
+
"step": 24600
|
2176 |
+
},
|
2177 |
+
{
|
2178 |
+
"epoch": 46.69,
|
2179 |
+
"learning_rate": 5e-05,
|
2180 |
+
"loss": 6.9053,
|
2181 |
+
"step": 24700
|
2182 |
+
},
|
2183 |
+
{
|
2184 |
+
"epoch": 46.88,
|
2185 |
+
"learning_rate": 5e-05,
|
2186 |
+
"loss": 6.8891,
|
2187 |
+
"step": 24800
|
2188 |
+
},
|
2189 |
+
{
|
2190 |
+
"epoch": 47.0,
|
2191 |
+
"eval_accuracy": 0.22235897435897437,
|
2192 |
+
"eval_loss": 6.312758922576904,
|
2193 |
+
"eval_runtime": 4.4993,
|
2194 |
+
"eval_samples_per_second": 111.128,
|
2195 |
+
"eval_steps_per_second": 14.002,
|
2196 |
+
"step": 24863
|
2197 |
+
},
|
2198 |
+
{
|
2199 |
+
"epoch": 47.0,
|
2200 |
+
"eval_exact_match": 0.0,
|
2201 |
+
"eval_f1": 0.0,
|
2202 |
+
"step": 24863
|
2203 |
+
},
|
2204 |
+
{
|
2205 |
+
"epoch": 47.07,
|
2206 |
+
"learning_rate": 5e-05,
|
2207 |
+
"loss": 6.8946,
|
2208 |
+
"step": 24900
|
2209 |
+
},
|
2210 |
+
{
|
2211 |
+
"epoch": 47.26,
|
2212 |
+
"learning_rate": 5e-05,
|
2213 |
+
"loss": 6.891,
|
2214 |
+
"step": 25000
|
2215 |
+
},
|
2216 |
+
{
|
2217 |
+
"epoch": 47.45,
|
2218 |
+
"learning_rate": 5e-05,
|
2219 |
+
"loss": 6.8629,
|
2220 |
+
"step": 25100
|
2221 |
+
},
|
2222 |
+
{
|
2223 |
+
"epoch": 47.64,
|
2224 |
+
"learning_rate": 5e-05,
|
2225 |
+
"loss": 6.8565,
|
2226 |
+
"step": 25200
|
2227 |
+
},
|
2228 |
+
{
|
2229 |
+
"epoch": 47.83,
|
2230 |
+
"learning_rate": 5e-05,
|
2231 |
+
"loss": 6.8621,
|
2232 |
+
"step": 25300
|
2233 |
+
},
|
2234 |
+
{
|
2235 |
+
"epoch": 48.0,
|
2236 |
+
"eval_accuracy": 0.23,
|
2237 |
+
"eval_loss": 6.2464728355407715,
|
2238 |
+
"eval_runtime": 4.2093,
|
2239 |
+
"eval_samples_per_second": 118.783,
|
2240 |
+
"eval_steps_per_second": 14.967,
|
2241 |
+
"step": 25392
|
2242 |
+
},
|
2243 |
+
{
|
2244 |
+
"epoch": 48.0,
|
2245 |
+
"eval_exact_match": 0.0,
|
2246 |
+
"eval_f1": 0.0,
|
2247 |
+
"step": 25392
|
2248 |
+
},
|
2249 |
+
{
|
2250 |
+
"epoch": 48.02,
|
2251 |
+
"learning_rate": 5e-05,
|
2252 |
+
"loss": 6.8663,
|
2253 |
+
"step": 25400
|
2254 |
+
},
|
2255 |
+
{
|
2256 |
+
"epoch": 48.2,
|
2257 |
+
"learning_rate": 5e-05,
|
2258 |
+
"loss": 6.8607,
|
2259 |
+
"step": 25500
|
2260 |
+
},
|
2261 |
+
{
|
2262 |
+
"epoch": 48.39,
|
2263 |
+
"learning_rate": 5e-05,
|
2264 |
+
"loss": 6.8698,
|
2265 |
+
"step": 25600
|
2266 |
+
},
|
2267 |
+
{
|
2268 |
+
"epoch": 48.58,
|
2269 |
+
"learning_rate": 5e-05,
|
2270 |
+
"loss": 6.8366,
|
2271 |
+
"step": 25700
|
2272 |
+
},
|
2273 |
+
{
|
2274 |
+
"epoch": 48.77,
|
2275 |
+
"learning_rate": 5e-05,
|
2276 |
+
"loss": 6.8415,
|
2277 |
+
"step": 25800
|
2278 |
+
},
|
2279 |
+
{
|
2280 |
+
"epoch": 48.96,
|
2281 |
+
"learning_rate": 5e-05,
|
2282 |
+
"loss": 6.8176,
|
2283 |
+
"step": 25900
|
2284 |
+
},
|
2285 |
+
{
|
2286 |
+
"epoch": 49.0,
|
2287 |
+
"eval_accuracy": 0.2561025641025641,
|
2288 |
+
"eval_loss": 6.196552753448486,
|
2289 |
+
"eval_runtime": 4.411,
|
2290 |
+
"eval_samples_per_second": 113.353,
|
2291 |
+
"eval_steps_per_second": 14.282,
|
2292 |
+
"step": 25921
|
2293 |
+
},
|
2294 |
+
{
|
2295 |
+
"epoch": 49.0,
|
2296 |
+
"eval_exact_match": 0.0,
|
2297 |
+
"eval_f1": 0.0,
|
2298 |
+
"step": 25921
|
2299 |
+
},
|
2300 |
+
{
|
2301 |
+
"epoch": 49.15,
|
2302 |
+
"learning_rate": 5e-05,
|
2303 |
+
"loss": 6.8197,
|
2304 |
+
"step": 26000
|
2305 |
+
},
|
2306 |
+
{
|
2307 |
+
"epoch": 49.34,
|
2308 |
+
"learning_rate": 5e-05,
|
2309 |
+
"loss": 6.8248,
|
2310 |
+
"step": 26100
|
2311 |
+
},
|
2312 |
+
{
|
2313 |
+
"epoch": 49.53,
|
2314 |
+
"learning_rate": 5e-05,
|
2315 |
+
"loss": 6.8394,
|
2316 |
+
"step": 26200
|
2317 |
+
},
|
2318 |
+
{
|
2319 |
+
"epoch": 49.72,
|
2320 |
+
"learning_rate": 5e-05,
|
2321 |
+
"loss": 6.7962,
|
2322 |
+
"step": 26300
|
2323 |
+
},
|
2324 |
+
{
|
2325 |
+
"epoch": 49.91,
|
2326 |
+
"learning_rate": 5e-05,
|
2327 |
+
"loss": 6.9473,
|
2328 |
+
"step": 26400
|
2329 |
+
},
|
2330 |
+
{
|
2331 |
+
"epoch": 50.0,
|
2332 |
+
"eval_accuracy": 0.19302564102564101,
|
2333 |
+
"eval_loss": 6.791760444641113,
|
2334 |
+
"eval_runtime": 4.1966,
|
2335 |
+
"eval_samples_per_second": 119.145,
|
2336 |
+
"eval_steps_per_second": 15.012,
|
2337 |
+
"step": 26450
|
2338 |
+
},
|
2339 |
+
{
|
2340 |
+
"epoch": 50.0,
|
2341 |
+
"eval_exact_match": 0.0,
|
2342 |
+
"eval_f1": 0.0,
|
2343 |
+
"step": 26450
|
2344 |
+
},
|
2345 |
+
{
|
2346 |
+
"epoch": 50.0,
|
2347 |
+
"step": 26450,
|
2348 |
+
"total_flos": 4.466691005730521e+18,
|
2349 |
+
"train_loss": 4.017823467146472,
|
2350 |
+
"train_runtime": 26175.2328,
|
2351 |
+
"train_samples_per_second": 32.33,
|
2352 |
+
"train_steps_per_second": 1.01
|
2353 |
+
}
|
2354 |
+
],
|
2355 |
+
"logging_steps": 100,
|
2356 |
+
"max_steps": 26450,
|
2357 |
+
"num_train_epochs": 50,
|
2358 |
+
"save_steps": 500,
|
2359 |
+
"total_flos": 4.466691005730521e+18,
|
2360 |
+
"trial_name": null,
|
2361 |
+
"trial_params": null
|
2362 |
+
}
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4728
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:02ea6908168bff114859f50f96da6b6e0a0236be9ba4f244f08a94b9ed4fe63b
|
3 |
size 4728
|