Add 2233188
Browse files- README.md +49 -4
- all_results.json +5 -5
- pytorch_model.bin +1 -1
- train_results.json +5 -5
- trainer_state.json +588 -3
README.md
CHANGED
|
@@ -11,9 +11,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 11 |
|
| 12 |
# out
|
| 13 |
|
| 14 |
-
This model is a fine-tuned version of [/1TB_SSD/SB_AI/
|
| 15 |
It achieves the following results on the evaluation set:
|
| 16 |
-
- Loss: 0.
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
|
|
@@ -35,10 +35,10 @@ The following hyperparameters were used during training:
|
|
| 35 |
- learning_rate: 5e-05
|
| 36 |
- train_batch_size: 1
|
| 37 |
- eval_batch_size: 1
|
| 38 |
-
- seed:
|
| 39 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 40 |
- lr_scheduler_type: linear
|
| 41 |
-
- num_epochs:
|
| 42 |
|
| 43 |
### Training results
|
| 44 |
|
|
@@ -58,6 +58,51 @@ The following hyperparameters were used during training:
|
|
| 58 |
| 0.059 | 0.81 | 900000 | 0.0652 |
|
| 59 |
| 0.0666 | 0.87 | 975000 | 0.0619 |
|
| 60 |
| 0.0624 | 0.94 | 1050000 | 0.0619 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 61 |
|
| 62 |
|
| 63 |
### Framework versions
|
|
|
|
| 11 |
|
| 12 |
# out
|
| 13 |
|
| 14 |
+
This model is a fine-tuned version of [/1TB_SSD/SB_AI/out_epoch1/out/checkpoint-1115000/](https://huggingface.co//1TB_SSD/SB_AI/out_epoch1/out/checkpoint-1115000/) on an unknown dataset.
|
| 15 |
It achieves the following results on the evaluation set:
|
| 16 |
+
- Loss: 0.0645
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
|
|
|
|
| 35 |
- learning_rate: 5e-05
|
| 36 |
- train_batch_size: 1
|
| 37 |
- eval_batch_size: 1
|
| 38 |
+
- seed: 2518227880
|
| 39 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 40 |
- lr_scheduler_type: linear
|
| 41 |
+
- num_epochs: 2.0
|
| 42 |
|
| 43 |
### Training results
|
| 44 |
|
|
|
|
| 58 |
| 0.059 | 0.81 | 900000 | 0.0652 |
|
| 59 |
| 0.0666 | 0.87 | 975000 | 0.0619 |
|
| 60 |
| 0.0624 | 0.94 | 1050000 | 0.0619 |
|
| 61 |
+
| 0.0625 | 1.01 | 1125000 | 0.0667 |
|
| 62 |
+
| 0.0614 | 1.03 | 1150000 | 0.0658 |
|
| 63 |
+
| 0.0597 | 1.05 | 1175000 | 0.0683 |
|
| 64 |
+
| 0.0629 | 1.07 | 1200000 | 0.0691 |
|
| 65 |
+
| 0.0603 | 1.1 | 1225000 | 0.0678 |
|
| 66 |
+
| 0.0601 | 1.12 | 1250000 | 0.0746 |
|
| 67 |
+
| 0.0606 | 1.14 | 1275000 | 0.0691 |
|
| 68 |
+
| 0.0671 | 1.16 | 1300000 | 0.0702 |
|
| 69 |
+
| 0.0625 | 1.19 | 1325000 | 0.0661 |
|
| 70 |
+
| 0.0617 | 1.21 | 1350000 | 0.0688 |
|
| 71 |
+
| 0.0579 | 1.23 | 1375000 | 0.0679 |
|
| 72 |
+
| 0.0663 | 1.25 | 1400000 | 0.0634 |
|
| 73 |
+
| 0.0583 | 1.28 | 1425000 | 0.0638 |
|
| 74 |
+
| 0.0623 | 1.3 | 1450000 | 0.0681 |
|
| 75 |
+
| 0.0615 | 1.32 | 1475000 | 0.0670 |
|
| 76 |
+
| 0.0592 | 1.34 | 1500000 | 0.0666 |
|
| 77 |
+
| 0.0626 | 1.37 | 1525000 | 0.0666 |
|
| 78 |
+
| 0.063 | 1.39 | 1550000 | 0.0647 |
|
| 79 |
+
| 0.0648 | 1.41 | 1575000 | 0.0653 |
|
| 80 |
+
| 0.0611 | 1.43 | 1600000 | 0.0700 |
|
| 81 |
+
| 0.0622 | 1.46 | 1625000 | 0.0634 |
|
| 82 |
+
| 0.0617 | 1.48 | 1650000 | 0.0651 |
|
| 83 |
+
| 0.0613 | 1.5 | 1675000 | 0.0634 |
|
| 84 |
+
| 0.0639 | 1.52 | 1700000 | 0.0661 |
|
| 85 |
+
| 0.0615 | 1.54 | 1725000 | 0.0644 |
|
| 86 |
+
| 0.0605 | 1.57 | 1750000 | 0.0662 |
|
| 87 |
+
| 0.0622 | 1.59 | 1775000 | 0.0656 |
|
| 88 |
+
| 0.0585 | 1.61 | 1800000 | 0.0633 |
|
| 89 |
+
| 0.0628 | 1.63 | 1825000 | 0.0625 |
|
| 90 |
+
| 0.0638 | 1.66 | 1850000 | 0.0662 |
|
| 91 |
+
| 0.0599 | 1.68 | 1875000 | 0.0664 |
|
| 92 |
+
| 0.0583 | 1.7 | 1900000 | 0.0668 |
|
| 93 |
+
| 0.0543 | 1.72 | 1925000 | 0.0631 |
|
| 94 |
+
| 0.06 | 1.75 | 1950000 | 0.0629 |
|
| 95 |
+
| 0.0615 | 1.77 | 1975000 | 0.0644 |
|
| 96 |
+
| 0.0587 | 1.79 | 2000000 | 0.0663 |
|
| 97 |
+
| 0.0647 | 1.81 | 2025000 | 0.0654 |
|
| 98 |
+
| 0.0604 | 1.84 | 2050000 | 0.0639 |
|
| 99 |
+
| 0.0641 | 1.86 | 2075000 | 0.0636 |
|
| 100 |
+
| 0.0604 | 1.88 | 2100000 | 0.0636 |
|
| 101 |
+
| 0.0654 | 1.9 | 2125000 | 0.0652 |
|
| 102 |
+
| 0.0588 | 1.93 | 2150000 | 0.0638 |
|
| 103 |
+
| 0.0616 | 1.95 | 2175000 | 0.0657 |
|
| 104 |
+
| 0.0598 | 1.97 | 2200000 | 0.0646 |
|
| 105 |
+
| 0.0633 | 1.99 | 2225000 | 0.0645 |
|
| 106 |
|
| 107 |
|
| 108 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 1116594,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"train_loss": 0.030777697094727392,
|
| 4 |
+
"train_runtime": 228453.6417,
|
| 5 |
"train_samples": 1116594,
|
| 6 |
+
"train_samples_per_second": 9.775,
|
| 7 |
+
"train_steps_per_second": 9.775
|
| 8 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 891703231
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1d0ec9b60b45e3f2d6bbd55005877c63954154cb113d9575bd472932dc918566
|
| 3 |
size 891703231
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 1116594,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"train_loss": 0.030777697094727392,
|
| 4 |
+
"train_runtime": 228453.6417,
|
| 5 |
"train_samples": 1116594,
|
| 6 |
+
"train_samples_per_second": 9.775,
|
| 7 |
+
"train_steps_per_second": 9.775
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -2578,11 +2578,596 @@
|
|
| 2578 |
"eval_samples_per_second": 36.912,
|
| 2579 |
"eval_steps_per_second": 36.912,
|
| 2580 |
"step": 1850000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2581 |
}
|
| 2582 |
],
|
| 2583 |
"max_steps": 2233188,
|
| 2584 |
"num_train_epochs": 2,
|
| 2585 |
-
"total_flos":
|
| 2586 |
"trial_name": null,
|
| 2587 |
"trial_params": null
|
| 2588 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"global_step": 2233188,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 2578 |
"eval_samples_per_second": 36.912,
|
| 2579 |
"eval_steps_per_second": 36.912,
|
| 2580 |
"step": 1850000
|
| 2581 |
+
},
|
| 2582 |
+
{
|
| 2583 |
+
"epoch": 1.66,
|
| 2584 |
+
"learning_rate": 8.467446538312046e-06,
|
| 2585 |
+
"loss": 0.0601,
|
| 2586 |
+
"step": 1855000
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 1.67,
|
| 2590 |
+
"learning_rate": 8.355498954857361e-06,
|
| 2591 |
+
"loss": 0.065,
|
| 2592 |
+
"step": 1860000
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 1.67,
|
| 2596 |
+
"learning_rate": 8.243551371402678e-06,
|
| 2597 |
+
"loss": 0.059,
|
| 2598 |
+
"step": 1865000
|
| 2599 |
+
},
|
| 2600 |
+
{
|
| 2601 |
+
"epoch": 1.67,
|
| 2602 |
+
"learning_rate": 8.131603787947991e-06,
|
| 2603 |
+
"loss": 0.064,
|
| 2604 |
+
"step": 1870000
|
| 2605 |
+
},
|
| 2606 |
+
{
|
| 2607 |
+
"epoch": 1.68,
|
| 2608 |
+
"learning_rate": 8.019656204493308e-06,
|
| 2609 |
+
"loss": 0.0599,
|
| 2610 |
+
"step": 1875000
|
| 2611 |
+
},
|
| 2612 |
+
{
|
| 2613 |
+
"epoch": 1.68,
|
| 2614 |
+
"eval_loss": 0.06644026190042496,
|
| 2615 |
+
"eval_runtime": 1678.6507,
|
| 2616 |
+
"eval_samples_per_second": 36.954,
|
| 2617 |
+
"eval_steps_per_second": 36.954,
|
| 2618 |
+
"step": 1875000
|
| 2619 |
+
},
|
| 2620 |
+
{
|
| 2621 |
+
"epoch": 1.68,
|
| 2622 |
+
"learning_rate": 7.907708621038623e-06,
|
| 2623 |
+
"loss": 0.0639,
|
| 2624 |
+
"step": 1880000
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 1.69,
|
| 2628 |
+
"learning_rate": 7.795761037583939e-06,
|
| 2629 |
+
"loss": 0.0631,
|
| 2630 |
+
"step": 1885000
|
| 2631 |
+
},
|
| 2632 |
+
{
|
| 2633 |
+
"epoch": 1.69,
|
| 2634 |
+
"learning_rate": 7.683813454129254e-06,
|
| 2635 |
+
"loss": 0.064,
|
| 2636 |
+
"step": 1890000
|
| 2637 |
+
},
|
| 2638 |
+
{
|
| 2639 |
+
"epoch": 1.7,
|
| 2640 |
+
"learning_rate": 7.57186587067457e-06,
|
| 2641 |
+
"loss": 0.062,
|
| 2642 |
+
"step": 1895000
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 1.7,
|
| 2646 |
+
"learning_rate": 7.459918287219884e-06,
|
| 2647 |
+
"loss": 0.0583,
|
| 2648 |
+
"step": 1900000
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 1.7,
|
| 2652 |
+
"eval_loss": 0.0668446272611618,
|
| 2653 |
+
"eval_runtime": 1681.2378,
|
| 2654 |
+
"eval_samples_per_second": 36.897,
|
| 2655 |
+
"eval_steps_per_second": 36.897,
|
| 2656 |
+
"step": 1900000
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 1.71,
|
| 2660 |
+
"learning_rate": 7.3479707037652e-06,
|
| 2661 |
+
"loss": 0.0599,
|
| 2662 |
+
"step": 1905000
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 1.71,
|
| 2666 |
+
"learning_rate": 7.236023120310516e-06,
|
| 2667 |
+
"loss": 0.0611,
|
| 2668 |
+
"step": 1910000
|
| 2669 |
+
},
|
| 2670 |
+
{
|
| 2671 |
+
"epoch": 1.72,
|
| 2672 |
+
"learning_rate": 7.124075536855831e-06,
|
| 2673 |
+
"loss": 0.0627,
|
| 2674 |
+
"step": 1915000
|
| 2675 |
+
},
|
| 2676 |
+
{
|
| 2677 |
+
"epoch": 1.72,
|
| 2678 |
+
"learning_rate": 7.012127953401147e-06,
|
| 2679 |
+
"loss": 0.0591,
|
| 2680 |
+
"step": 1920000
|
| 2681 |
+
},
|
| 2682 |
+
{
|
| 2683 |
+
"epoch": 1.72,
|
| 2684 |
+
"learning_rate": 6.900180369946463e-06,
|
| 2685 |
+
"loss": 0.0543,
|
| 2686 |
+
"step": 1925000
|
| 2687 |
+
},
|
| 2688 |
+
{
|
| 2689 |
+
"epoch": 1.72,
|
| 2690 |
+
"eval_loss": 0.06305810809135437,
|
| 2691 |
+
"eval_runtime": 1683.6555,
|
| 2692 |
+
"eval_samples_per_second": 36.844,
|
| 2693 |
+
"eval_steps_per_second": 36.844,
|
| 2694 |
+
"step": 1925000
|
| 2695 |
+
},
|
| 2696 |
+
{
|
| 2697 |
+
"epoch": 1.73,
|
| 2698 |
+
"learning_rate": 6.788232786491779e-06,
|
| 2699 |
+
"loss": 0.0622,
|
| 2700 |
+
"step": 1930000
|
| 2701 |
+
},
|
| 2702 |
+
{
|
| 2703 |
+
"epoch": 1.73,
|
| 2704 |
+
"learning_rate": 6.676285203037093e-06,
|
| 2705 |
+
"loss": 0.0605,
|
| 2706 |
+
"step": 1935000
|
| 2707 |
+
},
|
| 2708 |
+
{
|
| 2709 |
+
"epoch": 1.74,
|
| 2710 |
+
"learning_rate": 6.564337619582408e-06,
|
| 2711 |
+
"loss": 0.0622,
|
| 2712 |
+
"step": 1940000
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 1.74,
|
| 2716 |
+
"learning_rate": 6.452390036127724e-06,
|
| 2717 |
+
"loss": 0.0596,
|
| 2718 |
+
"step": 1945000
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 1.75,
|
| 2722 |
+
"learning_rate": 6.34044245267304e-06,
|
| 2723 |
+
"loss": 0.06,
|
| 2724 |
+
"step": 1950000
|
| 2725 |
+
},
|
| 2726 |
+
{
|
| 2727 |
+
"epoch": 1.75,
|
| 2728 |
+
"eval_loss": 0.06286083161830902,
|
| 2729 |
+
"eval_runtime": 1684.2136,
|
| 2730 |
+
"eval_samples_per_second": 36.832,
|
| 2731 |
+
"eval_steps_per_second": 36.832,
|
| 2732 |
+
"step": 1950000
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 1.75,
|
| 2736 |
+
"learning_rate": 6.2284948692183554e-06,
|
| 2737 |
+
"loss": 0.058,
|
| 2738 |
+
"step": 1955000
|
| 2739 |
+
},
|
| 2740 |
+
{
|
| 2741 |
+
"epoch": 1.76,
|
| 2742 |
+
"learning_rate": 6.116547285763671e-06,
|
| 2743 |
+
"loss": 0.0594,
|
| 2744 |
+
"step": 1960000
|
| 2745 |
+
},
|
| 2746 |
+
{
|
| 2747 |
+
"epoch": 1.76,
|
| 2748 |
+
"learning_rate": 6.004599702308987e-06,
|
| 2749 |
+
"loss": 0.0679,
|
| 2750 |
+
"step": 1965000
|
| 2751 |
+
},
|
| 2752 |
+
{
|
| 2753 |
+
"epoch": 1.76,
|
| 2754 |
+
"learning_rate": 5.892652118854302e-06,
|
| 2755 |
+
"loss": 0.0621,
|
| 2756 |
+
"step": 1970000
|
| 2757 |
+
},
|
| 2758 |
+
{
|
| 2759 |
+
"epoch": 1.77,
|
| 2760 |
+
"learning_rate": 5.780704535399617e-06,
|
| 2761 |
+
"loss": 0.0615,
|
| 2762 |
+
"step": 1975000
|
| 2763 |
+
},
|
| 2764 |
+
{
|
| 2765 |
+
"epoch": 1.77,
|
| 2766 |
+
"eval_loss": 0.06435712426900864,
|
| 2767 |
+
"eval_runtime": 1684.3427,
|
| 2768 |
+
"eval_samples_per_second": 36.829,
|
| 2769 |
+
"eval_steps_per_second": 36.829,
|
| 2770 |
+
"step": 1975000
|
| 2771 |
+
},
|
| 2772 |
+
{
|
| 2773 |
+
"epoch": 1.77,
|
| 2774 |
+
"learning_rate": 5.668756951944933e-06,
|
| 2775 |
+
"loss": 0.0602,
|
| 2776 |
+
"step": 1980000
|
| 2777 |
+
},
|
| 2778 |
+
{
|
| 2779 |
+
"epoch": 1.78,
|
| 2780 |
+
"learning_rate": 5.556809368490248e-06,
|
| 2781 |
+
"loss": 0.0589,
|
| 2782 |
+
"step": 1985000
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 1.78,
|
| 2786 |
+
"learning_rate": 5.444861785035564e-06,
|
| 2787 |
+
"loss": 0.0578,
|
| 2788 |
+
"step": 1990000
|
| 2789 |
+
},
|
| 2790 |
+
{
|
| 2791 |
+
"epoch": 1.79,
|
| 2792 |
+
"learning_rate": 5.332914201580879e-06,
|
| 2793 |
+
"loss": 0.0626,
|
| 2794 |
+
"step": 1995000
|
| 2795 |
+
},
|
| 2796 |
+
{
|
| 2797 |
+
"epoch": 1.79,
|
| 2798 |
+
"learning_rate": 5.220966618126195e-06,
|
| 2799 |
+
"loss": 0.0587,
|
| 2800 |
+
"step": 2000000
|
| 2801 |
+
},
|
| 2802 |
+
{
|
| 2803 |
+
"epoch": 1.79,
|
| 2804 |
+
"eval_loss": 0.06628864258527756,
|
| 2805 |
+
"eval_runtime": 1682.6389,
|
| 2806 |
+
"eval_samples_per_second": 36.866,
|
| 2807 |
+
"eval_steps_per_second": 36.866,
|
| 2808 |
+
"step": 2000000
|
| 2809 |
+
},
|
| 2810 |
+
{
|
| 2811 |
+
"epoch": 1.8,
|
| 2812 |
+
"learning_rate": 5.109019034671511e-06,
|
| 2813 |
+
"loss": 0.0595,
|
| 2814 |
+
"step": 2005000
|
| 2815 |
+
},
|
| 2816 |
+
{
|
| 2817 |
+
"epoch": 1.8,
|
| 2818 |
+
"learning_rate": 4.997071451216825e-06,
|
| 2819 |
+
"loss": 0.0584,
|
| 2820 |
+
"step": 2010000
|
| 2821 |
+
},
|
| 2822 |
+
{
|
| 2823 |
+
"epoch": 1.8,
|
| 2824 |
+
"learning_rate": 4.885123867762141e-06,
|
| 2825 |
+
"loss": 0.0562,
|
| 2826 |
+
"step": 2015000
|
| 2827 |
+
},
|
| 2828 |
+
{
|
| 2829 |
+
"epoch": 1.81,
|
| 2830 |
+
"learning_rate": 4.773176284307457e-06,
|
| 2831 |
+
"loss": 0.0694,
|
| 2832 |
+
"step": 2020000
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 1.81,
|
| 2836 |
+
"learning_rate": 4.661228700852772e-06,
|
| 2837 |
+
"loss": 0.0647,
|
| 2838 |
+
"step": 2025000
|
| 2839 |
+
},
|
| 2840 |
+
{
|
| 2841 |
+
"epoch": 1.81,
|
| 2842 |
+
"eval_loss": 0.06544966250658035,
|
| 2843 |
+
"eval_runtime": 1682.8498,
|
| 2844 |
+
"eval_samples_per_second": 36.862,
|
| 2845 |
+
"eval_steps_per_second": 36.862,
|
| 2846 |
+
"step": 2025000
|
| 2847 |
+
},
|
| 2848 |
+
{
|
| 2849 |
+
"epoch": 1.82,
|
| 2850 |
+
"learning_rate": 4.5492811173980875e-06,
|
| 2851 |
+
"loss": 0.0596,
|
| 2852 |
+
"step": 2030000
|
| 2853 |
+
},
|
| 2854 |
+
{
|
| 2855 |
+
"epoch": 1.82,
|
| 2856 |
+
"learning_rate": 4.4373335339434035e-06,
|
| 2857 |
+
"loss": 0.062,
|
| 2858 |
+
"step": 2035000
|
| 2859 |
+
},
|
| 2860 |
+
{
|
| 2861 |
+
"epoch": 1.83,
|
| 2862 |
+
"learning_rate": 4.325385950488719e-06,
|
| 2863 |
+
"loss": 0.0656,
|
| 2864 |
+
"step": 2040000
|
| 2865 |
+
},
|
| 2866 |
+
{
|
| 2867 |
+
"epoch": 1.83,
|
| 2868 |
+
"learning_rate": 4.213438367034034e-06,
|
| 2869 |
+
"loss": 0.0632,
|
| 2870 |
+
"step": 2045000
|
| 2871 |
+
},
|
| 2872 |
+
{
|
| 2873 |
+
"epoch": 1.84,
|
| 2874 |
+
"learning_rate": 4.101490783579349e-06,
|
| 2875 |
+
"loss": 0.0604,
|
| 2876 |
+
"step": 2050000
|
| 2877 |
+
},
|
| 2878 |
+
{
|
| 2879 |
+
"epoch": 1.84,
|
| 2880 |
+
"eval_loss": 0.06386958807706833,
|
| 2881 |
+
"eval_runtime": 1684.8253,
|
| 2882 |
+
"eval_samples_per_second": 36.819,
|
| 2883 |
+
"eval_steps_per_second": 36.819,
|
| 2884 |
+
"step": 2050000
|
| 2885 |
+
},
|
| 2886 |
+
{
|
| 2887 |
+
"epoch": 1.84,
|
| 2888 |
+
"learning_rate": 3.989543200124665e-06,
|
| 2889 |
+
"loss": 0.0586,
|
| 2890 |
+
"step": 2055000
|
| 2891 |
+
},
|
| 2892 |
+
{
|
| 2893 |
+
"epoch": 1.84,
|
| 2894 |
+
"learning_rate": 3.87759561666998e-06,
|
| 2895 |
+
"loss": 0.0595,
|
| 2896 |
+
"step": 2060000
|
| 2897 |
+
},
|
| 2898 |
+
{
|
| 2899 |
+
"epoch": 1.85,
|
| 2900 |
+
"learning_rate": 3.765648033215296e-06,
|
| 2901 |
+
"loss": 0.0628,
|
| 2902 |
+
"step": 2065000
|
| 2903 |
+
},
|
| 2904 |
+
{
|
| 2905 |
+
"epoch": 1.85,
|
| 2906 |
+
"learning_rate": 3.6537004497606115e-06,
|
| 2907 |
+
"loss": 0.0614,
|
| 2908 |
+
"step": 2070000
|
| 2909 |
+
},
|
| 2910 |
+
{
|
| 2911 |
+
"epoch": 1.86,
|
| 2912 |
+
"learning_rate": 3.541752866305927e-06,
|
| 2913 |
+
"loss": 0.0641,
|
| 2914 |
+
"step": 2075000
|
| 2915 |
+
},
|
| 2916 |
+
{
|
| 2917 |
+
"epoch": 1.86,
|
| 2918 |
+
"eval_loss": 0.06358933448791504,
|
| 2919 |
+
"eval_runtime": 1682.4468,
|
| 2920 |
+
"eval_samples_per_second": 36.871,
|
| 2921 |
+
"eval_steps_per_second": 36.871,
|
| 2922 |
+
"step": 2075000
|
| 2923 |
+
},
|
| 2924 |
+
{
|
| 2925 |
+
"epoch": 1.86,
|
| 2926 |
+
"learning_rate": 3.4298052828512423e-06,
|
| 2927 |
+
"loss": 0.0612,
|
| 2928 |
+
"step": 2080000
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 1.87,
|
| 2932 |
+
"learning_rate": 3.317857699396558e-06,
|
| 2933 |
+
"loss": 0.0675,
|
| 2934 |
+
"step": 2085000
|
| 2935 |
+
},
|
| 2936 |
+
{
|
| 2937 |
+
"epoch": 1.87,
|
| 2938 |
+
"learning_rate": 3.2059101159418736e-06,
|
| 2939 |
+
"loss": 0.0657,
|
| 2940 |
+
"step": 2090000
|
| 2941 |
+
},
|
| 2942 |
+
{
|
| 2943 |
+
"epoch": 1.88,
|
| 2944 |
+
"learning_rate": 3.0939625324871888e-06,
|
| 2945 |
+
"loss": 0.0628,
|
| 2946 |
+
"step": 2095000
|
| 2947 |
+
},
|
| 2948 |
+
{
|
| 2949 |
+
"epoch": 1.88,
|
| 2950 |
+
"learning_rate": 2.9820149490325044e-06,
|
| 2951 |
+
"loss": 0.0604,
|
| 2952 |
+
"step": 2100000
|
| 2953 |
+
},
|
| 2954 |
+
{
|
| 2955 |
+
"epoch": 1.88,
|
| 2956 |
+
"eval_loss": 0.0636412724852562,
|
| 2957 |
+
"eval_runtime": 1687.8932,
|
| 2958 |
+
"eval_samples_per_second": 36.752,
|
| 2959 |
+
"eval_steps_per_second": 36.752,
|
| 2960 |
+
"step": 2100000
|
| 2961 |
+
},
|
| 2962 |
+
{
|
| 2963 |
+
"epoch": 1.89,
|
| 2964 |
+
"learning_rate": 2.87006736557782e-06,
|
| 2965 |
+
"loss": 0.0592,
|
| 2966 |
+
"step": 2105000
|
| 2967 |
+
},
|
| 2968 |
+
{
|
| 2969 |
+
"epoch": 1.89,
|
| 2970 |
+
"learning_rate": 2.758119782123135e-06,
|
| 2971 |
+
"loss": 0.0619,
|
| 2972 |
+
"step": 2110000
|
| 2973 |
+
},
|
| 2974 |
+
{
|
| 2975 |
+
"epoch": 1.89,
|
| 2976 |
+
"learning_rate": 2.6461721986684508e-06,
|
| 2977 |
+
"loss": 0.0623,
|
| 2978 |
+
"step": 2115000
|
| 2979 |
+
},
|
| 2980 |
+
{
|
| 2981 |
+
"epoch": 1.9,
|
| 2982 |
+
"learning_rate": 2.5342246152137664e-06,
|
| 2983 |
+
"loss": 0.0568,
|
| 2984 |
+
"step": 2120000
|
| 2985 |
+
},
|
| 2986 |
+
{
|
| 2987 |
+
"epoch": 1.9,
|
| 2988 |
+
"learning_rate": 2.4222770317590816e-06,
|
| 2989 |
+
"loss": 0.0654,
|
| 2990 |
+
"step": 2125000
|
| 2991 |
+
},
|
| 2992 |
+
{
|
| 2993 |
+
"epoch": 1.9,
|
| 2994 |
+
"eval_loss": 0.06523974984884262,
|
| 2995 |
+
"eval_runtime": 1681.6119,
|
| 2996 |
+
"eval_samples_per_second": 36.889,
|
| 2997 |
+
"eval_steps_per_second": 36.889,
|
| 2998 |
+
"step": 2125000
|
| 2999 |
+
},
|
| 3000 |
+
{
|
| 3001 |
+
"epoch": 1.91,
|
| 3002 |
+
"learning_rate": 2.310329448304397e-06,
|
| 3003 |
+
"loss": 0.0597,
|
| 3004 |
+
"step": 2130000
|
| 3005 |
+
},
|
| 3006 |
+
{
|
| 3007 |
+
"epoch": 1.91,
|
| 3008 |
+
"learning_rate": 2.198381864849713e-06,
|
| 3009 |
+
"loss": 0.0614,
|
| 3010 |
+
"step": 2135000
|
| 3011 |
+
},
|
| 3012 |
+
{
|
| 3013 |
+
"epoch": 1.92,
|
| 3014 |
+
"learning_rate": 2.086434281395028e-06,
|
| 3015 |
+
"loss": 0.0576,
|
| 3016 |
+
"step": 2140000
|
| 3017 |
+
},
|
| 3018 |
+
{
|
| 3019 |
+
"epoch": 1.92,
|
| 3020 |
+
"learning_rate": 1.9744866979403436e-06,
|
| 3021 |
+
"loss": 0.0625,
|
| 3022 |
+
"step": 2145000
|
| 3023 |
+
},
|
| 3024 |
+
{
|
| 3025 |
+
"epoch": 1.93,
|
| 3026 |
+
"learning_rate": 1.862539114485659e-06,
|
| 3027 |
+
"loss": 0.0588,
|
| 3028 |
+
"step": 2150000
|
| 3029 |
+
},
|
| 3030 |
+
{
|
| 3031 |
+
"epoch": 1.93,
|
| 3032 |
+
"eval_loss": 0.06383997201919556,
|
| 3033 |
+
"eval_runtime": 1685.504,
|
| 3034 |
+
"eval_samples_per_second": 36.804,
|
| 3035 |
+
"eval_steps_per_second": 36.804,
|
| 3036 |
+
"step": 2150000
|
| 3037 |
+
},
|
| 3038 |
+
{
|
| 3039 |
+
"epoch": 1.93,
|
| 3040 |
+
"learning_rate": 1.7505915310309748e-06,
|
| 3041 |
+
"loss": 0.0525,
|
| 3042 |
+
"step": 2155000
|
| 3043 |
+
},
|
| 3044 |
+
{
|
| 3045 |
+
"epoch": 1.93,
|
| 3046 |
+
"learning_rate": 1.63864394757629e-06,
|
| 3047 |
+
"loss": 0.0573,
|
| 3048 |
+
"step": 2160000
|
| 3049 |
+
},
|
| 3050 |
+
{
|
| 3051 |
+
"epoch": 1.94,
|
| 3052 |
+
"learning_rate": 1.5266963641216056e-06,
|
| 3053 |
+
"loss": 0.0612,
|
| 3054 |
+
"step": 2165000
|
| 3055 |
+
},
|
| 3056 |
+
{
|
| 3057 |
+
"epoch": 1.94,
|
| 3058 |
+
"learning_rate": 1.4147487806669212e-06,
|
| 3059 |
+
"loss": 0.0625,
|
| 3060 |
+
"step": 2170000
|
| 3061 |
+
},
|
| 3062 |
+
{
|
| 3063 |
+
"epoch": 1.95,
|
| 3064 |
+
"learning_rate": 1.3028011972122364e-06,
|
| 3065 |
+
"loss": 0.0616,
|
| 3066 |
+
"step": 2175000
|
| 3067 |
+
},
|
| 3068 |
+
{
|
| 3069 |
+
"epoch": 1.95,
|
| 3070 |
+
"eval_loss": 0.06566296517848969,
|
| 3071 |
+
"eval_runtime": 1686.3029,
|
| 3072 |
+
"eval_samples_per_second": 36.786,
|
| 3073 |
+
"eval_steps_per_second": 36.786,
|
| 3074 |
+
"step": 2175000
|
| 3075 |
+
},
|
| 3076 |
+
{
|
| 3077 |
+
"epoch": 1.95,
|
| 3078 |
+
"learning_rate": 1.190853613757552e-06,
|
| 3079 |
+
"loss": 0.0629,
|
| 3080 |
+
"step": 2180000
|
| 3081 |
+
},
|
| 3082 |
+
{
|
| 3083 |
+
"epoch": 1.96,
|
| 3084 |
+
"learning_rate": 1.0789060303028674e-06,
|
| 3085 |
+
"loss": 0.0543,
|
| 3086 |
+
"step": 2185000
|
| 3087 |
+
},
|
| 3088 |
+
{
|
| 3089 |
+
"epoch": 1.96,
|
| 3090 |
+
"learning_rate": 9.66958446848183e-07,
|
| 3091 |
+
"loss": 0.0584,
|
| 3092 |
+
"step": 2190000
|
| 3093 |
+
},
|
| 3094 |
+
{
|
| 3095 |
+
"epoch": 1.97,
|
| 3096 |
+
"learning_rate": 8.550108633934986e-07,
|
| 3097 |
+
"loss": 0.0595,
|
| 3098 |
+
"step": 2195000
|
| 3099 |
+
},
|
| 3100 |
+
{
|
| 3101 |
+
"epoch": 1.97,
|
| 3102 |
+
"learning_rate": 7.43063279938814e-07,
|
| 3103 |
+
"loss": 0.0598,
|
| 3104 |
+
"step": 2200000
|
| 3105 |
+
},
|
| 3106 |
+
{
|
| 3107 |
+
"epoch": 1.97,
|
| 3108 |
+
"eval_loss": 0.06455818563699722,
|
| 3109 |
+
"eval_runtime": 1682.3023,
|
| 3110 |
+
"eval_samples_per_second": 36.874,
|
| 3111 |
+
"eval_steps_per_second": 36.874,
|
| 3112 |
+
"step": 2200000
|
| 3113 |
+
},
|
| 3114 |
+
{
|
| 3115 |
+
"epoch": 1.97,
|
| 3116 |
+
"learning_rate": 6.311156964841294e-07,
|
| 3117 |
+
"loss": 0.0595,
|
| 3118 |
+
"step": 2205000
|
| 3119 |
+
},
|
| 3120 |
+
{
|
| 3121 |
+
"epoch": 1.98,
|
| 3122 |
+
"learning_rate": 5.191681130294449e-07,
|
| 3123 |
+
"loss": 0.0602,
|
| 3124 |
+
"step": 2210000
|
| 3125 |
+
},
|
| 3126 |
+
{
|
| 3127 |
+
"epoch": 1.98,
|
| 3128 |
+
"learning_rate": 4.0722052957476043e-07,
|
| 3129 |
+
"loss": 0.0601,
|
| 3130 |
+
"step": 2215000
|
| 3131 |
+
},
|
| 3132 |
+
{
|
| 3133 |
+
"epoch": 1.99,
|
| 3134 |
+
"learning_rate": 2.952729461200759e-07,
|
| 3135 |
+
"loss": 0.0595,
|
| 3136 |
+
"step": 2220000
|
| 3137 |
+
},
|
| 3138 |
+
{
|
| 3139 |
+
"epoch": 1.99,
|
| 3140 |
+
"learning_rate": 1.8332536266539136e-07,
|
| 3141 |
+
"loss": 0.0633,
|
| 3142 |
+
"step": 2225000
|
| 3143 |
+
},
|
| 3144 |
+
{
|
| 3145 |
+
"epoch": 1.99,
|
| 3146 |
+
"eval_loss": 0.06449371576309204,
|
| 3147 |
+
"eval_runtime": 1682.7954,
|
| 3148 |
+
"eval_samples_per_second": 36.863,
|
| 3149 |
+
"eval_steps_per_second": 36.863,
|
| 3150 |
+
"step": 2225000
|
| 3151 |
+
},
|
| 3152 |
+
{
|
| 3153 |
+
"epoch": 2.0,
|
| 3154 |
+
"learning_rate": 7.137777921070686e-08,
|
| 3155 |
+
"loss": 0.0603,
|
| 3156 |
+
"step": 2230000
|
| 3157 |
+
},
|
| 3158 |
+
{
|
| 3159 |
+
"epoch": 2.0,
|
| 3160 |
+
"step": 2233188,
|
| 3161 |
+
"total_flos": 6.706494317713766e+17,
|
| 3162 |
+
"train_loss": 0.030777697094727392,
|
| 3163 |
+
"train_runtime": 228453.6417,
|
| 3164 |
+
"train_samples_per_second": 9.775,
|
| 3165 |
+
"train_steps_per_second": 9.775
|
| 3166 |
}
|
| 3167 |
],
|
| 3168 |
"max_steps": 2233188,
|
| 3169 |
"num_train_epochs": 2,
|
| 3170 |
+
"total_flos": 6.706494317713766e+17,
|
| 3171 |
"trial_name": null,
|
| 3172 |
"trial_params": null
|
| 3173 |
}
|