Commit
•
8e62039
1
Parent(s):
c4bcd05
Model save
Browse files- README.md +61 -14
- all_results.json +5 -5
- pytorch_model.bin +1 -1
- train_results.json +5 -5
- trainer_state.json +741 -36
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -7,7 +7,7 @@ datasets:
|
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
| 9 |
model-index:
|
| 10 |
-
- name:
|
| 11 |
results:
|
| 12 |
- task:
|
| 13 |
name: Causal Language Modeling
|
|
@@ -18,18 +18,18 @@ model-index:
|
|
| 18 |
metrics:
|
| 19 |
- name: Accuracy
|
| 20 |
type: accuracy
|
| 21 |
-
value: 0.
|
| 22 |
---
|
| 23 |
|
| 24 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 25 |
should probably proofread and complete it, then remove this comment. -->
|
| 26 |
|
| 27 |
-
#
|
| 28 |
|
| 29 |
This model is a fine-tuned version of [facebook/opt-1.3b](https://huggingface.co/facebook/opt-1.3b) on the AlekseyKorshuk/dalio-handwritten-io dataset.
|
| 30 |
It achieves the following results on the evaluation set:
|
| 31 |
-
- Loss: 2.
|
| 32 |
-
- Accuracy: 0.
|
| 33 |
|
| 34 |
## Model description
|
| 35 |
|
|
@@ -49,24 +49,71 @@ More information needed
|
|
| 49 |
|
| 50 |
The following hyperparameters were used during training:
|
| 51 |
- learning_rate: 3e-05
|
| 52 |
-
- train_batch_size:
|
| 53 |
-
- eval_batch_size:
|
| 54 |
- seed: 42
|
| 55 |
- distributed_type: multi-GPU
|
| 56 |
- num_devices: 8
|
| 57 |
-
- total_train_batch_size:
|
| 58 |
-
- total_eval_batch_size:
|
| 59 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 60 |
-
- lr_scheduler_type:
|
| 61 |
-
- num_epochs:
|
| 62 |
|
| 63 |
### Training results
|
| 64 |
|
| 65 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 66 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 67 |
-
| 2.
|
| 68 |
-
| 2.
|
| 69 |
-
| 2.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 70 |
|
| 71 |
|
| 72 |
### Framework versions
|
|
|
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
| 9 |
model-index:
|
| 10 |
+
- name: test-clm
|
| 11 |
results:
|
| 12 |
- task:
|
| 13 |
name: Causal Language Modeling
|
|
|
|
| 18 |
metrics:
|
| 19 |
- name: Accuracy
|
| 20 |
type: accuracy
|
| 21 |
+
value: 0.06414321574844761
|
| 22 |
---
|
| 23 |
|
| 24 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 25 |
should probably proofread and complete it, then remove this comment. -->
|
| 26 |
|
| 27 |
+
# test-clm
|
| 28 |
|
| 29 |
This model is a fine-tuned version of [facebook/opt-1.3b](https://huggingface.co/facebook/opt-1.3b) on the AlekseyKorshuk/dalio-handwritten-io dataset.
|
| 30 |
It achieves the following results on the evaluation set:
|
| 31 |
+
- Loss: 2.5547
|
| 32 |
+
- Accuracy: 0.0641
|
| 33 |
|
| 34 |
## Model description
|
| 35 |
|
|
|
|
| 49 |
|
| 50 |
The following hyperparameters were used during training:
|
| 51 |
- learning_rate: 3e-05
|
| 52 |
+
- train_batch_size: 2
|
| 53 |
+
- eval_batch_size: 2
|
| 54 |
- seed: 42
|
| 55 |
- distributed_type: multi-GPU
|
| 56 |
- num_devices: 8
|
| 57 |
+
- total_train_batch_size: 16
|
| 58 |
+
- total_eval_batch_size: 16
|
| 59 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 60 |
+
- lr_scheduler_type: cosine
|
| 61 |
+
- num_epochs: 5.0
|
| 62 |
|
| 63 |
### Training results
|
| 64 |
|
| 65 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 66 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 67 |
+
| 2.9219 | 0.1 | 1 | 2.6484 | 0.0529 |
|
| 68 |
+
| 2.6938 | 0.2 | 2 | 2.6484 | 0.0529 |
|
| 69 |
+
| 2.6365 | 0.3 | 3 | 2.5508 | 0.0560 |
|
| 70 |
+
| 2.5088 | 0.4 | 4 | 2.5332 | 0.0563 |
|
| 71 |
+
| 2.7297 | 0.5 | 5 | 2.5176 | 0.0567 |
|
| 72 |
+
| 2.9702 | 0.6 | 6 | 2.4941 | 0.0572 |
|
| 73 |
+
| 2.729 | 0.7 | 7 | 2.4883 | 0.0568 |
|
| 74 |
+
| 2.6172 | 0.8 | 8 | 2.4785 | 0.0578 |
|
| 75 |
+
| 2.6428 | 0.9 | 9 | 2.4590 | 0.0581 |
|
| 76 |
+
| 2.5681 | 1.0 | 10 | 2.4355 | 0.0590 |
|
| 77 |
+
| 2.1885 | 1.1 | 11 | 2.4238 | 0.0587 |
|
| 78 |
+
| 1.981 | 1.2 | 12 | 2.4219 | 0.0587 |
|
| 79 |
+
| 1.8673 | 1.3 | 13 | 2.4180 | 0.0591 |
|
| 80 |
+
| 1.7321 | 1.4 | 14 | 2.4180 | 0.0596 |
|
| 81 |
+
| 1.6355 | 1.5 | 15 | 2.4180 | 0.0601 |
|
| 82 |
+
| 1.7758 | 1.6 | 16 | 2.4199 | 0.0602 |
|
| 83 |
+
| 2.0162 | 1.7 | 17 | 2.4082 | 0.0605 |
|
| 84 |
+
| 1.8037 | 1.8 | 18 | 2.3965 | 0.0605 |
|
| 85 |
+
| 1.7204 | 1.9 | 19 | 2.375 | 0.0608 |
|
| 86 |
+
| 1.7831 | 2.0 | 20 | 2.3574 | 0.0609 |
|
| 87 |
+
| 1.299 | 2.1 | 21 | 2.3496 | 0.0616 |
|
| 88 |
+
| 1.4463 | 2.2 | 22 | 2.3496 | 0.0620 |
|
| 89 |
+
| 1.1733 | 2.3 | 23 | 2.3652 | 0.0617 |
|
| 90 |
+
| 1.1142 | 2.4 | 24 | 2.3887 | 0.0626 |
|
| 91 |
+
| 1.3107 | 2.5 | 25 | 2.4219 | 0.0627 |
|
| 92 |
+
| 1.011 | 2.6 | 26 | 2.4551 | 0.0622 |
|
| 93 |
+
| 1.3403 | 2.7 | 27 | 2.4766 | 0.0616 |
|
| 94 |
+
| 1.3108 | 2.8 | 28 | 2.4766 | 0.0616 |
|
| 95 |
+
| 1.0076 | 2.9 | 29 | 2.4609 | 0.0619 |
|
| 96 |
+
| 0.8656 | 3.0 | 30 | 2.4512 | 0.0624 |
|
| 97 |
+
| 0.6635 | 3.1 | 31 | 2.4512 | 0.0628 |
|
| 98 |
+
| 0.9996 | 3.2 | 32 | 2.4434 | 0.0635 |
|
| 99 |
+
| 0.9029 | 3.3 | 33 | 2.4473 | 0.0637 |
|
| 100 |
+
| 0.8329 | 3.4 | 34 | 2.4551 | 0.0637 |
|
| 101 |
+
| 0.8012 | 3.5 | 35 | 2.4648 | 0.0639 |
|
| 102 |
+
| 0.5814 | 3.6 | 36 | 2.4902 | 0.0640 |
|
| 103 |
+
| 1.0688 | 3.7 | 37 | 2.5098 | 0.0638 |
|
| 104 |
+
| 0.8688 | 3.8 | 38 | 2.5176 | 0.0635 |
|
| 105 |
+
| 0.7341 | 3.9 | 39 | 2.5195 | 0.0638 |
|
| 106 |
+
| 0.7102 | 4.0 | 40 | 2.5195 | 0.0640 |
|
| 107 |
+
| 0.7079 | 4.1 | 41 | 2.5195 | 0.0641 |
|
| 108 |
+
| 0.7656 | 4.2 | 42 | 2.5195 | 0.0643 |
|
| 109 |
+
| 0.6377 | 4.3 | 43 | 2.5273 | 0.0645 |
|
| 110 |
+
| 0.5898 | 4.4 | 44 | 2.5352 | 0.0641 |
|
| 111 |
+
| 0.5958 | 4.5 | 45 | 2.5430 | 0.0641 |
|
| 112 |
+
| 0.7048 | 4.6 | 46 | 2.5488 | 0.0640 |
|
| 113 |
+
| 0.5435 | 4.7 | 47 | 2.5527 | 0.0641 |
|
| 114 |
+
| 0.4769 | 4.8 | 48 | 2.5527 | 0.0640 |
|
| 115 |
+
| 0.6583 | 4.9 | 49 | 2.5547 | 0.0642 |
|
| 116 |
+
| 0.7168 | 5.0 | 50 | 2.5547 | 0.0641 |
|
| 117 |
|
| 118 |
|
| 119 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 156,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 1.4392181396484376,
|
| 4 |
+
"train_runtime": 183.8573,
|
| 5 |
"train_samples": 156,
|
| 6 |
+
"train_samples_per_second": 4.242,
|
| 7 |
+
"train_steps_per_second": 0.272
|
| 8 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2631610617
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6206eb9213367136975c480c7b17e06148f1ce3be73c7b78ca2c2373ce011857
|
| 3 |
size 2631610617
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 156,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 1.4392181396484376,
|
| 4 |
+
"train_runtime": 183.8573,
|
| 5 |
"train_samples": 156,
|
| 6 |
+
"train_samples_per_second": 4.242,
|
| 7 |
+
"train_steps_per_second": 0.272
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,70 +1,775 @@
|
|
| 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,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
-
"epoch": 0.
|
| 12 |
"learning_rate": 3e-05,
|
| 13 |
-
"loss": 2.
|
| 14 |
"step": 1
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 0.
|
| 18 |
-
"eval_accuracy": 0.
|
| 19 |
-
"eval_loss": 2.
|
| 20 |
-
"eval_runtime": 1.
|
| 21 |
-
"eval_samples_per_second":
|
| 22 |
-
"eval_steps_per_second":
|
| 23 |
"step": 1
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
"learning_rate": 3e-05,
|
| 28 |
-
"loss": 2.
|
| 29 |
"step": 2
|
| 30 |
},
|
| 31 |
{
|
| 32 |
-
"epoch": 0.
|
| 33 |
-
"eval_accuracy": 0.
|
| 34 |
-
"eval_loss": 2.
|
| 35 |
-
"eval_runtime": 1.
|
| 36 |
-
"eval_samples_per_second":
|
| 37 |
-
"eval_steps_per_second":
|
| 38 |
"step": 2
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 2.
|
| 44 |
"step": 3
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch":
|
| 48 |
-
"eval_accuracy": 0.
|
| 49 |
-
"eval_loss": 2.
|
| 50 |
-
"eval_runtime": 1.
|
| 51 |
-
"eval_samples_per_second":
|
| 52 |
-
"eval_steps_per_second":
|
| 53 |
"step": 3
|
| 54 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
{
|
| 56 |
"epoch": 1.0,
|
| 57 |
-
"
|
| 58 |
-
"
|
| 59 |
-
"
|
| 60 |
-
"
|
| 61 |
-
"
|
| 62 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 63 |
}
|
| 64 |
],
|
| 65 |
-
"max_steps":
|
| 66 |
-
"num_train_epochs":
|
| 67 |
-
"total_flos":
|
| 68 |
"trial_name": null,
|
| 69 |
"trial_params": null
|
| 70 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 50,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.1,
|
| 12 |
"learning_rate": 3e-05,
|
| 13 |
+
"loss": 2.9219,
|
| 14 |
"step": 1
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 0.1,
|
| 18 |
+
"eval_accuracy": 0.05291319857312723,
|
| 19 |
+
"eval_loss": 2.6484375,
|
| 20 |
+
"eval_runtime": 1.5576,
|
| 21 |
+
"eval_samples_per_second": 18.618,
|
| 22 |
+
"eval_steps_per_second": 1.284,
|
| 23 |
"step": 1
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.2,
|
| 27 |
"learning_rate": 3e-05,
|
| 28 |
+
"loss": 2.6938,
|
| 29 |
"step": 2
|
| 30 |
},
|
| 31 |
{
|
| 32 |
+
"epoch": 0.2,
|
| 33 |
+
"eval_accuracy": 0.05291319857312723,
|
| 34 |
+
"eval_loss": 2.6484375,
|
| 35 |
+
"eval_runtime": 1.275,
|
| 36 |
+
"eval_samples_per_second": 22.746,
|
| 37 |
+
"eval_steps_per_second": 1.569,
|
| 38 |
"step": 2
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.3,
|
| 42 |
+
"learning_rate": 2.9970400926424075e-05,
|
| 43 |
+
"loss": 2.6365,
|
| 44 |
"step": 3
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.3,
|
| 48 |
+
"eval_accuracy": 0.05601796802748051,
|
| 49 |
+
"eval_loss": 2.55078125,
|
| 50 |
+
"eval_runtime": 1.1095,
|
| 51 |
+
"eval_samples_per_second": 26.137,
|
| 52 |
+
"eval_steps_per_second": 1.803,
|
| 53 |
"step": 3
|
| 54 |
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.4,
|
| 57 |
+
"learning_rate": 2.988172051971717e-05,
|
| 58 |
+
"loss": 2.5088,
|
| 59 |
+
"step": 4
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.4,
|
| 63 |
+
"eval_accuracy": 0.056282203725723345,
|
| 64 |
+
"eval_loss": 2.533203125,
|
| 65 |
+
"eval_runtime": 1.0069,
|
| 66 |
+
"eval_samples_per_second": 28.8,
|
| 67 |
+
"eval_steps_per_second": 1.986,
|
| 68 |
+
"step": 4
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.5,
|
| 72 |
+
"learning_rate": 2.9734308760930333e-05,
|
| 73 |
+
"loss": 2.7297,
|
| 74 |
+
"step": 5
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.5,
|
| 78 |
+
"eval_accuracy": 0.05667855727308759,
|
| 79 |
+
"eval_loss": 2.517578125,
|
| 80 |
+
"eval_runtime": 1.0127,
|
| 81 |
+
"eval_samples_per_second": 28.637,
|
| 82 |
+
"eval_steps_per_second": 1.975,
|
| 83 |
+
"step": 5
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.6,
|
| 87 |
+
"learning_rate": 2.9528747416929467e-05,
|
| 88 |
+
"loss": 2.9702,
|
| 89 |
+
"step": 6
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.6,
|
| 93 |
+
"eval_accuracy": 0.05720702866957326,
|
| 94 |
+
"eval_loss": 2.494140625,
|
| 95 |
+
"eval_runtime": 1.2159,
|
| 96 |
+
"eval_samples_per_second": 23.851,
|
| 97 |
+
"eval_steps_per_second": 1.645,
|
| 98 |
+
"step": 6
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.7,
|
| 102 |
+
"learning_rate": 2.9265847744427305e-05,
|
| 103 |
+
"loss": 2.729,
|
| 104 |
+
"step": 7
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.7,
|
| 108 |
+
"eval_accuracy": 0.05681067512220901,
|
| 109 |
+
"eval_loss": 2.48828125,
|
| 110 |
+
"eval_runtime": 1.0262,
|
| 111 |
+
"eval_samples_per_second": 28.259,
|
| 112 |
+
"eval_steps_per_second": 1.949,
|
| 113 |
+
"step": 7
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.8,
|
| 117 |
+
"learning_rate": 2.894664728832377e-05,
|
| 118 |
+
"loss": 2.6172,
|
| 119 |
+
"step": 8
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.8,
|
| 123 |
+
"eval_accuracy": 0.05780155899061963,
|
| 124 |
+
"eval_loss": 2.478515625,
|
| 125 |
+
"eval_runtime": 1.3127,
|
| 126 |
+
"eval_samples_per_second": 22.092,
|
| 127 |
+
"eval_steps_per_second": 1.524,
|
| 128 |
+
"step": 8
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.9,
|
| 132 |
+
"learning_rate": 2.8572405786990293e-05,
|
| 133 |
+
"loss": 2.6428,
|
| 134 |
+
"step": 9
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.9,
|
| 138 |
+
"eval_accuracy": 0.058065794688862464,
|
| 139 |
+
"eval_loss": 2.458984375,
|
| 140 |
+
"eval_runtime": 1.0971,
|
| 141 |
+
"eval_samples_per_second": 26.432,
|
| 142 |
+
"eval_steps_per_second": 1.823,
|
| 143 |
+
"step": 9
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.0,
|
| 147 |
+
"learning_rate": 2.8144600200657953e-05,
|
| 148 |
+
"loss": 2.5681,
|
| 149 |
+
"step": 10
|
| 150 |
+
},
|
| 151 |
{
|
| 152 |
"epoch": 1.0,
|
| 153 |
+
"eval_accuracy": 0.05899061963271238,
|
| 154 |
+
"eval_loss": 2.435546875,
|
| 155 |
+
"eval_runtime": 1.3171,
|
| 156 |
+
"eval_samples_per_second": 22.018,
|
| 157 |
+
"eval_steps_per_second": 1.518,
|
| 158 |
+
"step": 10
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.1,
|
| 162 |
+
"learning_rate": 2.7664918882530227e-05,
|
| 163 |
+
"loss": 2.1885,
|
| 164 |
+
"step": 11
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.1,
|
| 168 |
+
"eval_accuracy": 0.05866032500990884,
|
| 169 |
+
"eval_loss": 2.423828125,
|
| 170 |
+
"eval_runtime": 1.2181,
|
| 171 |
+
"eval_samples_per_second": 23.807,
|
| 172 |
+
"eval_steps_per_second": 1.642,
|
| 173 |
+
"step": 11
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 1.2,
|
| 177 |
+
"learning_rate": 2.7135254915624213e-05,
|
| 178 |
+
"loss": 1.981,
|
| 179 |
+
"step": 12
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.2,
|
| 183 |
+
"eval_accuracy": 0.05872638393446954,
|
| 184 |
+
"eval_loss": 2.421875,
|
| 185 |
+
"eval_runtime": 1.6039,
|
| 186 |
+
"eval_samples_per_second": 18.081,
|
| 187 |
+
"eval_steps_per_second": 1.247,
|
| 188 |
+
"step": 12
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.3,
|
| 192 |
+
"learning_rate": 2.655769864163684e-05,
|
| 193 |
+
"loss": 1.8673,
|
| 194 |
+
"step": 13
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.3,
|
| 198 |
+
"eval_accuracy": 0.0591227374818338,
|
| 199 |
+
"eval_loss": 2.41796875,
|
| 200 |
+
"eval_runtime": 1.0123,
|
| 201 |
+
"eval_samples_per_second": 28.649,
|
| 202 |
+
"eval_steps_per_second": 1.976,
|
| 203 |
+
"step": 13
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 1.4,
|
| 207 |
+
"learning_rate": 2.5934529411321174e-05,
|
| 208 |
+
"loss": 1.7321,
|
| 209 |
+
"step": 14
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.4,
|
| 213 |
+
"eval_accuracy": 0.05958514995375875,
|
| 214 |
+
"eval_loss": 2.41796875,
|
| 215 |
+
"eval_runtime": 1.0152,
|
| 216 |
+
"eval_samples_per_second": 28.567,
|
| 217 |
+
"eval_steps_per_second": 1.97,
|
| 218 |
+
"step": 14
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.5,
|
| 222 |
+
"learning_rate": 2.5268206588930332e-05,
|
| 223 |
+
"loss": 1.6355,
|
| 224 |
+
"step": 15
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.5,
|
| 228 |
+
"eval_accuracy": 0.060113621350244416,
|
| 229 |
+
"eval_loss": 2.41796875,
|
| 230 |
+
"eval_runtime": 1.4219,
|
| 231 |
+
"eval_samples_per_second": 20.396,
|
| 232 |
+
"eval_steps_per_second": 1.407,
|
| 233 |
+
"step": 15
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.6,
|
| 237 |
+
"learning_rate": 2.4561359846230346e-05,
|
| 238 |
+
"loss": 1.7758,
|
| 239 |
+
"step": 16
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 1.6,
|
| 243 |
+
"eval_accuracy": 0.06017968027480513,
|
| 244 |
+
"eval_loss": 2.419921875,
|
| 245 |
+
"eval_runtime": 1.618,
|
| 246 |
+
"eval_samples_per_second": 17.923,
|
| 247 |
+
"eval_steps_per_second": 1.236,
|
| 248 |
+
"step": 16
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.7,
|
| 252 |
+
"learning_rate": 2.3816778784387097e-05,
|
| 253 |
+
"loss": 2.0162,
|
| 254 |
+
"step": 17
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.7,
|
| 258 |
+
"eval_accuracy": 0.06050997489760867,
|
| 259 |
+
"eval_loss": 2.408203125,
|
| 260 |
+
"eval_runtime": 1.1225,
|
| 261 |
+
"eval_samples_per_second": 25.835,
|
| 262 |
+
"eval_steps_per_second": 1.782,
|
| 263 |
+
"step": 17
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.8,
|
| 267 |
+
"learning_rate": 2.303740192468495e-05,
|
| 268 |
+
"loss": 1.8037,
|
| 269 |
+
"step": 18
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.8,
|
| 273 |
+
"eval_accuracy": 0.06050997489760867,
|
| 274 |
+
"eval_loss": 2.396484375,
|
| 275 |
+
"eval_runtime": 0.912,
|
| 276 |
+
"eval_samples_per_second": 31.8,
|
| 277 |
+
"eval_steps_per_second": 2.193,
|
| 278 |
+
"step": 18
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 1.9,
|
| 282 |
+
"learning_rate": 2.222630511152573e-05,
|
| 283 |
+
"loss": 1.7204,
|
| 284 |
+
"step": 19
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.9,
|
| 288 |
+
"eval_accuracy": 0.0607742105958515,
|
| 289 |
+
"eval_loss": 2.375,
|
| 290 |
+
"eval_runtime": 1.2202,
|
| 291 |
+
"eval_samples_per_second": 23.767,
|
| 292 |
+
"eval_steps_per_second": 1.639,
|
| 293 |
+
"step": 19
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 2.0,
|
| 297 |
+
"learning_rate": 2.138668937347609e-05,
|
| 298 |
+
"loss": 1.7831,
|
| 299 |
+
"step": 20
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 2.0,
|
| 303 |
+
"eval_accuracy": 0.060906328444972915,
|
| 304 |
+
"eval_loss": 2.357421875,
|
| 305 |
+
"eval_runtime": 1.0082,
|
| 306 |
+
"eval_samples_per_second": 28.765,
|
| 307 |
+
"eval_steps_per_second": 1.984,
|
| 308 |
+
"step": 20
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 2.1,
|
| 312 |
+
"learning_rate": 2.052186829027017e-05,
|
| 313 |
+
"loss": 1.299,
|
| 314 |
+
"step": 21
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 2.1,
|
| 318 |
+
"eval_accuracy": 0.06163297661514071,
|
| 319 |
+
"eval_loss": 2.349609375,
|
| 320 |
+
"eval_runtime": 1.2114,
|
| 321 |
+
"eval_samples_per_second": 23.94,
|
| 322 |
+
"eval_steps_per_second": 1.651,
|
| 323 |
+
"step": 21
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 2.2,
|
| 327 |
+
"learning_rate": 1.963525491562421e-05,
|
| 328 |
+
"loss": 1.4463,
|
| 329 |
+
"step": 22
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 2.2,
|
| 333 |
+
"eval_accuracy": 0.06196327123794425,
|
| 334 |
+
"eval_loss": 2.349609375,
|
| 335 |
+
"eval_runtime": 1.5179,
|
| 336 |
+
"eval_samples_per_second": 19.105,
|
| 337 |
+
"eval_steps_per_second": 1.318,
|
| 338 |
+
"step": 22
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 2.3,
|
| 342 |
+
"learning_rate": 1.8730348307472828e-05,
|
| 343 |
+
"loss": 1.1733,
|
| 344 |
+
"step": 23
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 2.3,
|
| 348 |
+
"eval_accuracy": 0.061699035539701415,
|
| 349 |
+
"eval_loss": 2.365234375,
|
| 350 |
+
"eval_runtime": 1.3077,
|
| 351 |
+
"eval_samples_per_second": 22.176,
|
| 352 |
+
"eval_steps_per_second": 1.529,
|
| 353 |
+
"step": 23
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 2.4,
|
| 357 |
+
"learning_rate": 1.781071971878587e-05,
|
| 358 |
+
"loss": 1.1142,
|
| 359 |
+
"step": 24
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 2.4,
|
| 363 |
+
"eval_accuracy": 0.06255780155899061,
|
| 364 |
+
"eval_loss": 2.388671875,
|
| 365 |
+
"eval_runtime": 1.1042,
|
| 366 |
+
"eval_samples_per_second": 26.264,
|
| 367 |
+
"eval_steps_per_second": 1.811,
|
| 368 |
+
"step": 24
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.5,
|
| 372 |
+
"learning_rate": 1.6879998503464565e-05,
|
| 373 |
+
"loss": 1.3107,
|
| 374 |
+
"step": 25
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 2.5,
|
| 378 |
+
"eval_accuracy": 0.06268991940811204,
|
| 379 |
+
"eval_loss": 2.421875,
|
| 380 |
+
"eval_runtime": 0.9167,
|
| 381 |
+
"eval_samples_per_second": 31.634,
|
| 382 |
+
"eval_steps_per_second": 2.182,
|
| 383 |
+
"step": 25
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 2.6,
|
| 387 |
+
"learning_rate": 1.5941857792939702e-05,
|
| 388 |
+
"loss": 1.011,
|
| 389 |
+
"step": 26
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.6,
|
| 393 |
+
"eval_accuracy": 0.06216144801162637,
|
| 394 |
+
"eval_loss": 2.455078125,
|
| 395 |
+
"eval_runtime": 0.9207,
|
| 396 |
+
"eval_samples_per_second": 31.497,
|
| 397 |
+
"eval_steps_per_second": 2.172,
|
| 398 |
+
"step": 26
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 2.7,
|
| 402 |
+
"learning_rate": 1.5e-05,
|
| 403 |
+
"loss": 1.3403,
|
| 404 |
+
"step": 27
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 2.7,
|
| 408 |
+
"eval_accuracy": 0.061566917690579995,
|
| 409 |
+
"eval_loss": 2.4765625,
|
| 410 |
+
"eval_runtime": 1.5266,
|
| 411 |
+
"eval_samples_per_second": 18.997,
|
| 412 |
+
"eval_steps_per_second": 1.31,
|
| 413 |
+
"step": 27
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 2.8,
|
| 417 |
+
"learning_rate": 1.5e-05,
|
| 418 |
+
"loss": 1.3108,
|
| 419 |
+
"step": 28
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 2.8,
|
| 423 |
+
"eval_accuracy": 0.061566917690579995,
|
| 424 |
+
"eval_loss": 2.4765625,
|
| 425 |
+
"eval_runtime": 1.252,
|
| 426 |
+
"eval_samples_per_second": 23.163,
|
| 427 |
+
"eval_steps_per_second": 1.597,
|
| 428 |
+
"step": 28
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 2.9,
|
| 432 |
+
"learning_rate": 1.40581422070603e-05,
|
| 433 |
+
"loss": 1.0076,
|
| 434 |
+
"step": 29
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 2.9,
|
| 438 |
+
"eval_accuracy": 0.06189721231338354,
|
| 439 |
+
"eval_loss": 2.4609375,
|
| 440 |
+
"eval_runtime": 0.9112,
|
| 441 |
+
"eval_samples_per_second": 31.825,
|
| 442 |
+
"eval_steps_per_second": 2.195,
|
| 443 |
+
"step": 29
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 3.0,
|
| 447 |
+
"learning_rate": 1.3120001496535434e-05,
|
| 448 |
+
"loss": 0.8656,
|
| 449 |
+
"step": 30
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 3.0,
|
| 453 |
+
"eval_accuracy": 0.062359624785308494,
|
| 454 |
+
"eval_loss": 2.451171875,
|
| 455 |
+
"eval_runtime": 1.5156,
|
| 456 |
+
"eval_samples_per_second": 19.134,
|
| 457 |
+
"eval_steps_per_second": 1.32,
|
| 458 |
+
"step": 30
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 3.1,
|
| 462 |
+
"learning_rate": 1.2189280281214128e-05,
|
| 463 |
+
"loss": 0.6635,
|
| 464 |
+
"step": 31
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 3.1,
|
| 468 |
+
"eval_accuracy": 0.06282203725723345,
|
| 469 |
+
"eval_loss": 2.451171875,
|
| 470 |
+
"eval_runtime": 1.313,
|
| 471 |
+
"eval_samples_per_second": 22.087,
|
| 472 |
+
"eval_steps_per_second": 1.523,
|
| 473 |
+
"step": 31
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 3.2,
|
| 477 |
+
"learning_rate": 1.1269651692527181e-05,
|
| 478 |
+
"loss": 0.9996,
|
| 479 |
+
"step": 32
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 3.2,
|
| 483 |
+
"eval_accuracy": 0.06348262650284053,
|
| 484 |
+
"eval_loss": 2.443359375,
|
| 485 |
+
"eval_runtime": 1.01,
|
| 486 |
+
"eval_samples_per_second": 28.711,
|
| 487 |
+
"eval_steps_per_second": 1.98,
|
| 488 |
+
"step": 32
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 3.3,
|
| 492 |
+
"learning_rate": 1.036474508437579e-05,
|
| 493 |
+
"loss": 0.9029,
|
| 494 |
+
"step": 33
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 3.3,
|
| 498 |
+
"eval_accuracy": 0.06368080327652266,
|
| 499 |
+
"eval_loss": 2.447265625,
|
| 500 |
+
"eval_runtime": 1.4214,
|
| 501 |
+
"eval_samples_per_second": 20.402,
|
| 502 |
+
"eval_steps_per_second": 1.407,
|
| 503 |
+
"step": 33
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 3.4,
|
| 507 |
+
"learning_rate": 9.478131709729831e-06,
|
| 508 |
+
"loss": 0.8329,
|
| 509 |
+
"step": 34
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 3.4,
|
| 513 |
+
"eval_accuracy": 0.06374686220108336,
|
| 514 |
+
"eval_loss": 2.455078125,
|
| 515 |
+
"eval_runtime": 0.9136,
|
| 516 |
+
"eval_samples_per_second": 31.742,
|
| 517 |
+
"eval_steps_per_second": 2.189,
|
| 518 |
+
"step": 34
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 3.5,
|
| 522 |
+
"learning_rate": 8.61331062652391e-06,
|
| 523 |
+
"loss": 0.8012,
|
| 524 |
+
"step": 35
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 3.5,
|
| 528 |
+
"eval_accuracy": 0.06387898005020479,
|
| 529 |
+
"eval_loss": 2.46484375,
|
| 530 |
+
"eval_runtime": 1.6062,
|
| 531 |
+
"eval_samples_per_second": 18.055,
|
| 532 |
+
"eval_steps_per_second": 1.245,
|
| 533 |
+
"step": 35
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 3.6,
|
| 537 |
+
"learning_rate": 7.773694888474268e-06,
|
| 538 |
+
"loss": 0.5814,
|
| 539 |
+
"step": 36
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 3.6,
|
| 543 |
+
"eval_accuracy": 0.0640110978993262,
|
| 544 |
+
"eval_loss": 2.490234375,
|
| 545 |
+
"eval_runtime": 1.6209,
|
| 546 |
+
"eval_samples_per_second": 17.891,
|
| 547 |
+
"eval_steps_per_second": 1.234,
|
| 548 |
+
"step": 36
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 3.7,
|
| 552 |
+
"learning_rate": 6.962598075315047e-06,
|
| 553 |
+
"loss": 1.0688,
|
| 554 |
+
"step": 37
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 3.7,
|
| 558 |
+
"eval_accuracy": 0.06381292112564407,
|
| 559 |
+
"eval_loss": 2.509765625,
|
| 560 |
+
"eval_runtime": 1.2235,
|
| 561 |
+
"eval_samples_per_second": 23.703,
|
| 562 |
+
"eval_steps_per_second": 1.635,
|
| 563 |
+
"step": 37
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 3.8,
|
| 567 |
+
"learning_rate": 6.1832212156129045e-06,
|
| 568 |
+
"loss": 0.8688,
|
| 569 |
+
"step": 38
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 3.8,
|
| 573 |
+
"eval_accuracy": 0.06348262650284053,
|
| 574 |
+
"eval_loss": 2.517578125,
|
| 575 |
+
"eval_runtime": 1.5088,
|
| 576 |
+
"eval_samples_per_second": 19.221,
|
| 577 |
+
"eval_steps_per_second": 1.326,
|
| 578 |
+
"step": 38
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 3.9,
|
| 582 |
+
"learning_rate": 5.438640153769654e-06,
|
| 583 |
+
"loss": 0.7341,
|
| 584 |
+
"step": 39
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 3.9,
|
| 588 |
+
"eval_accuracy": 0.06381292112564407,
|
| 589 |
+
"eval_loss": 2.51953125,
|
| 590 |
+
"eval_runtime": 1.0209,
|
| 591 |
+
"eval_samples_per_second": 28.406,
|
| 592 |
+
"eval_steps_per_second": 1.959,
|
| 593 |
+
"step": 39
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 4.0,
|
| 597 |
+
"learning_rate": 4.731793411069669e-06,
|
| 598 |
+
"loss": 0.7102,
|
| 599 |
+
"step": 40
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 4.0,
|
| 603 |
+
"eval_accuracy": 0.0640110978993262,
|
| 604 |
+
"eval_loss": 2.51953125,
|
| 605 |
+
"eval_runtime": 1.3112,
|
| 606 |
+
"eval_samples_per_second": 22.116,
|
| 607 |
+
"eval_steps_per_second": 1.525,
|
| 608 |
+
"step": 40
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 4.1,
|
| 612 |
+
"learning_rate": 4.06547058867883e-06,
|
| 613 |
+
"loss": 0.7079,
|
| 614 |
+
"step": 41
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 4.1,
|
| 618 |
+
"eval_accuracy": 0.06414321574844761,
|
| 619 |
+
"eval_loss": 2.51953125,
|
| 620 |
+
"eval_runtime": 1.0046,
|
| 621 |
+
"eval_samples_per_second": 28.868,
|
| 622 |
+
"eval_steps_per_second": 1.991,
|
| 623 |
+
"step": 41
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 4.2,
|
| 627 |
+
"learning_rate": 3.442301358363163e-06,
|
| 628 |
+
"loss": 0.7656,
|
| 629 |
+
"step": 42
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 4.2,
|
| 633 |
+
"eval_accuracy": 0.06427533359756903,
|
| 634 |
+
"eval_loss": 2.51953125,
|
| 635 |
+
"eval_runtime": 1.3174,
|
| 636 |
+
"eval_samples_per_second": 22.012,
|
| 637 |
+
"eval_steps_per_second": 1.518,
|
| 638 |
+
"step": 42
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 4.3,
|
| 642 |
+
"learning_rate": 2.86474508437579e-06,
|
| 643 |
+
"loss": 0.6377,
|
| 644 |
+
"step": 43
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 4.3,
|
| 648 |
+
"eval_accuracy": 0.06447351037125115,
|
| 649 |
+
"eval_loss": 2.52734375,
|
| 650 |
+
"eval_runtime": 1.1043,
|
| 651 |
+
"eval_samples_per_second": 26.261,
|
| 652 |
+
"eval_steps_per_second": 1.811,
|
| 653 |
+
"step": 43
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 4.4,
|
| 657 |
+
"learning_rate": 2.335081117469777e-06,
|
| 658 |
+
"loss": 0.5898,
|
| 659 |
+
"step": 44
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 4.4,
|
| 663 |
+
"eval_accuracy": 0.06414321574844761,
|
| 664 |
+
"eval_loss": 2.53515625,
|
| 665 |
+
"eval_runtime": 1.0226,
|
| 666 |
+
"eval_samples_per_second": 28.359,
|
| 667 |
+
"eval_steps_per_second": 1.956,
|
| 668 |
+
"step": 44
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 4.5,
|
| 672 |
+
"learning_rate": 1.8553997993420495e-06,
|
| 673 |
+
"loss": 0.5958,
|
| 674 |
+
"step": 45
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 4.5,
|
| 678 |
+
"eval_accuracy": 0.06407715682388691,
|
| 679 |
+
"eval_loss": 2.54296875,
|
| 680 |
+
"eval_runtime": 0.91,
|
| 681 |
+
"eval_samples_per_second": 31.867,
|
| 682 |
+
"eval_steps_per_second": 2.198,
|
| 683 |
+
"step": 45
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 4.6,
|
| 687 |
+
"learning_rate": 1.4275942130097097e-06,
|
| 688 |
+
"loss": 0.7048,
|
| 689 |
+
"step": 46
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 4.6,
|
| 693 |
+
"eval_accuracy": 0.0640110978993262,
|
| 694 |
+
"eval_loss": 2.548828125,
|
| 695 |
+
"eval_runtime": 1.2089,
|
| 696 |
+
"eval_samples_per_second": 23.989,
|
| 697 |
+
"eval_steps_per_second": 1.654,
|
| 698 |
+
"step": 46
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 4.7,
|
| 702 |
+
"learning_rate": 1.0533527116762298e-06,
|
| 703 |
+
"loss": 0.5435,
|
| 704 |
+
"step": 47
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 4.7,
|
| 708 |
+
"eval_accuracy": 0.06414321574844761,
|
| 709 |
+
"eval_loss": 2.552734375,
|
| 710 |
+
"eval_runtime": 1.3311,
|
| 711 |
+
"eval_samples_per_second": 21.786,
|
| 712 |
+
"eval_steps_per_second": 1.502,
|
| 713 |
+
"step": 47
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 4.8,
|
| 717 |
+
"learning_rate": 7.341522555726971e-07,
|
| 718 |
+
"loss": 0.4769,
|
| 719 |
+
"step": 48
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 4.8,
|
| 723 |
+
"eval_accuracy": 0.0640110978993262,
|
| 724 |
+
"eval_loss": 2.552734375,
|
| 725 |
+
"eval_runtime": 1.0089,
|
| 726 |
+
"eval_samples_per_second": 28.743,
|
| 727 |
+
"eval_steps_per_second": 1.982,
|
| 728 |
+
"step": 48
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 4.9,
|
| 732 |
+
"learning_rate": 4.7125258307053385e-07,
|
| 733 |
+
"loss": 0.6583,
|
| 734 |
+
"step": 49
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 4.9,
|
| 738 |
+
"eval_accuracy": 0.06420927467300833,
|
| 739 |
+
"eval_loss": 2.5546875,
|
| 740 |
+
"eval_runtime": 1.0099,
|
| 741 |
+
"eval_samples_per_second": 28.715,
|
| 742 |
+
"eval_steps_per_second": 1.98,
|
| 743 |
+
"step": 49
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 5.0,
|
| 747 |
+
"learning_rate": 2.6569123906967083e-07,
|
| 748 |
+
"loss": 0.7168,
|
| 749 |
+
"step": 50
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 5.0,
|
| 753 |
+
"eval_accuracy": 0.06414321574844761,
|
| 754 |
+
"eval_loss": 2.5546875,
|
| 755 |
+
"eval_runtime": 1.5134,
|
| 756 |
+
"eval_samples_per_second": 19.163,
|
| 757 |
+
"eval_steps_per_second": 1.322,
|
| 758 |
+
"step": 50
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 5.0,
|
| 762 |
+
"step": 50,
|
| 763 |
+
"total_flos": 2477483753472.0,
|
| 764 |
+
"train_loss": 1.4392181396484376,
|
| 765 |
+
"train_runtime": 183.8573,
|
| 766 |
+
"train_samples_per_second": 4.242,
|
| 767 |
+
"train_steps_per_second": 0.272
|
| 768 |
}
|
| 769 |
],
|
| 770 |
+
"max_steps": 50,
|
| 771 |
+
"num_train_epochs": 5,
|
| 772 |
+
"total_flos": 2477483753472.0,
|
| 773 |
"trial_name": null,
|
| 774 |
"trial_params": null
|
| 775 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4591
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9567bff4e6027ed7913ec27f9c92249df3e60ca2394ff65e8c89def0feb5eac1
|
| 3 |
size 4591
|