Model save
Browse files- README.md +31 -19
- all_results.json +13 -0
- eval_results.json +8 -0
- model.safetensors +1 -1
- runs/Mar29_19-44-52_44990517b672/events.out.tfevents.1711742478.44990517b672.3784.2 +3 -0
- runs/Mar29_20-03-14_44990517b672/events.out.tfevents.1711742611.44990517b672.3784.3 +3 -0
- train_results.json +8 -0
- trainer_state.json +1329 -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 [google/vit-base-patch16-224-in21K](https://huggingface.co/google/vit-base-patch16-224-in21K) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -43,29 +43,41 @@ The following hyperparameters were used during training:
|
|
| 43 |
- seed: 42
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: linear
|
| 46 |
-
- num_epochs:
|
| 47 |
- mixed_precision_training: Native AMP
|
| 48 |
|
| 49 |
### Training results
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 52 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 53 |
-
|
|
| 54 |
-
|
|
| 55 |
-
|
|
| 56 |
-
|
|
| 57 |
-
|
|
| 58 |
-
|
|
| 59 |
-
|
|
| 60 |
-
|
|
| 61 |
-
|
|
| 62 |
-
|
|
| 63 |
-
|
|
| 64 |
-
|
|
| 65 |
-
|
|
| 66 |
-
|
|
| 67 |
-
|
|
| 68 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 69 |
|
| 70 |
|
| 71 |
### Framework versions
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21K](https://huggingface.co/google/vit-base-patch16-224-in21K) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.5081
|
| 21 |
+
- Accuracy: 0.8696
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 43 |
- seed: 42
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: linear
|
| 46 |
+
- num_epochs: 7
|
| 47 |
- mixed_precision_training: Native AMP
|
| 48 |
|
| 49 |
### Training results
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 52 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 53 |
+
| 1.3563 | 0.24 | 100 | 1.1495 | 0.6750 |
|
| 54 |
+
| 1.3393 | 0.48 | 200 | 1.0388 | 0.7204 |
|
| 55 |
+
| 1.2033 | 0.73 | 300 | 0.9324 | 0.7547 |
|
| 56 |
+
| 0.9672 | 0.97 | 400 | 0.8558 | 0.7659 |
|
| 57 |
+
| 0.8674 | 1.21 | 500 | 0.8456 | 0.7616 |
|
| 58 |
+
| 0.8277 | 1.45 | 600 | 0.7563 | 0.7959 |
|
| 59 |
+
| 0.8703 | 1.69 | 700 | 0.8465 | 0.7539 |
|
| 60 |
+
| 0.893 | 1.94 | 800 | 0.6881 | 0.8002 |
|
| 61 |
+
| 0.9454 | 2.18 | 900 | 0.7211 | 0.8027 |
|
| 62 |
+
| 0.8109 | 2.42 | 1000 | 0.6369 | 0.8285 |
|
| 63 |
+
| 0.8762 | 2.66 | 1100 | 0.6336 | 0.8396 |
|
| 64 |
+
| 0.8034 | 2.91 | 1200 | 0.6580 | 0.8165 |
|
| 65 |
+
| 0.5833 | 3.15 | 1300 | 0.5828 | 0.8439 |
|
| 66 |
+
| 0.8811 | 3.39 | 1400 | 0.6564 | 0.8259 |
|
| 67 |
+
| 0.5639 | 3.63 | 1500 | 0.5737 | 0.8439 |
|
| 68 |
+
| 0.639 | 3.87 | 1600 | 0.5609 | 0.8379 |
|
| 69 |
+
| 0.6455 | 4.12 | 1700 | 0.5820 | 0.8370 |
|
| 70 |
+
| 0.5402 | 4.36 | 1800 | 0.5797 | 0.8345 |
|
| 71 |
+
| 0.5311 | 4.6 | 1900 | 0.5511 | 0.8456 |
|
| 72 |
+
| 0.5734 | 4.84 | 2000 | 0.5444 | 0.8508 |
|
| 73 |
+
| 0.5206 | 5.08 | 2100 | 0.5326 | 0.8636 |
|
| 74 |
+
| 0.6272 | 5.33 | 2200 | 0.5478 | 0.8525 |
|
| 75 |
+
| 0.5124 | 5.57 | 2300 | 0.5296 | 0.8688 |
|
| 76 |
+
| 0.5659 | 5.81 | 2400 | 0.5181 | 0.8705 |
|
| 77 |
+
| 0.4212 | 6.05 | 2500 | 0.5200 | 0.8611 |
|
| 78 |
+
| 0.4338 | 6.3 | 2600 | 0.5135 | 0.8731 |
|
| 79 |
+
| 0.3407 | 6.54 | 2700 | 0.5147 | 0.8722 |
|
| 80 |
+
| 0.4043 | 6.78 | 2800 | 0.5081 | 0.8696 |
|
| 81 |
|
| 82 |
|
| 83 |
### Framework versions
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.7710120068610634,
|
| 4 |
+
"eval_loss": 0.912144124507904,
|
| 5 |
+
"eval_runtime": 6.7842,
|
| 6 |
+
"eval_samples_per_second": 171.871,
|
| 7 |
+
"eval_steps_per_second": 21.521,
|
| 8 |
+
"total_flos": 2.047635634195759e+18,
|
| 9 |
+
"train_loss": 1.8191430680543978,
|
| 10 |
+
"train_runtime": 594.0822,
|
| 11 |
+
"train_samples_per_second": 44.458,
|
| 12 |
+
"train_steps_per_second": 2.781
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.7710120068610634,
|
| 4 |
+
"eval_loss": 0.912144124507904,
|
| 5 |
+
"eval_runtime": 6.7842,
|
| 6 |
+
"eval_samples_per_second": 171.871,
|
| 7 |
+
"eval_steps_per_second": 21.521
|
| 8 |
+
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 343377784
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eb9b70ee282c426c2a645c30197368de15a5908c06f2c905de4194ed11a5c4ed
|
| 3 |
size 343377784
|
runs/Mar29_19-44-52_44990517b672/events.out.tfevents.1711742478.44990517b672.3784.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3636374ba0e100e35ba985b7def150b74b2d498b9b67d3dfddd2dc95add95a23
|
| 3 |
+
size 411
|
runs/Mar29_20-03-14_44990517b672/events.out.tfevents.1711742611.44990517b672.3784.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3c50db23ca1c40405186375018a2238b4a607bb59484fe5484ecc67db25f05dc
|
| 3 |
+
size 77831
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"total_flos": 2.047635634195759e+18,
|
| 4 |
+
"train_loss": 1.8191430680543978,
|
| 5 |
+
"train_runtime": 594.0822,
|
| 6 |
+
"train_samples_per_second": 44.458,
|
| 7 |
+
"train_steps_per_second": 2.781
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1329 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.912144124507904,
|
| 3 |
+
"best_model_checkpoint": "finetuned-cards-blackjack/checkpoint-1600",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 1652,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"grad_norm": 1.466597557067871,
|
| 14 |
+
"learning_rate": 0.0001987893462469734,
|
| 15 |
+
"loss": 3.9543,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05,
|
| 20 |
+
"grad_norm": 1.9476360082626343,
|
| 21 |
+
"learning_rate": 0.00019757869249394675,
|
| 22 |
+
"loss": 3.8868,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07,
|
| 27 |
+
"grad_norm": 1.6487232446670532,
|
| 28 |
+
"learning_rate": 0.0001963680387409201,
|
| 29 |
+
"loss": 3.8185,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.1,
|
| 34 |
+
"grad_norm": 1.8101606369018555,
|
| 35 |
+
"learning_rate": 0.00019515738498789345,
|
| 36 |
+
"loss": 3.6559,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.12,
|
| 41 |
+
"grad_norm": 1.7900973558425903,
|
| 42 |
+
"learning_rate": 0.00019394673123486684,
|
| 43 |
+
"loss": 3.559,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15,
|
| 48 |
+
"grad_norm": 1.7922214269638062,
|
| 49 |
+
"learning_rate": 0.0001927360774818402,
|
| 50 |
+
"loss": 3.4135,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.17,
|
| 55 |
+
"grad_norm": 1.9818700551986694,
|
| 56 |
+
"learning_rate": 0.00019152542372881357,
|
| 57 |
+
"loss": 3.3906,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.19,
|
| 62 |
+
"grad_norm": 1.9315565824508667,
|
| 63 |
+
"learning_rate": 0.00019031476997578695,
|
| 64 |
+
"loss": 3.3191,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.22,
|
| 69 |
+
"grad_norm": 1.9850099086761475,
|
| 70 |
+
"learning_rate": 0.0001891041162227603,
|
| 71 |
+
"loss": 3.2122,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.24,
|
| 76 |
+
"grad_norm": 1.9584887027740479,
|
| 77 |
+
"learning_rate": 0.00018789346246973366,
|
| 78 |
+
"loss": 3.1307,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.24,
|
| 83 |
+
"eval_accuracy": 0.20926243567753003,
|
| 84 |
+
"eval_loss": 3.017998456954956,
|
| 85 |
+
"eval_runtime": 6.19,
|
| 86 |
+
"eval_samples_per_second": 188.367,
|
| 87 |
+
"eval_steps_per_second": 23.586,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.27,
|
| 92 |
+
"grad_norm": 2.39744234085083,
|
| 93 |
+
"learning_rate": 0.00018668280871670702,
|
| 94 |
+
"loss": 3.0667,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.29,
|
| 99 |
+
"grad_norm": 1.891518473625183,
|
| 100 |
+
"learning_rate": 0.0001854721549636804,
|
| 101 |
+
"loss": 2.984,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.31,
|
| 106 |
+
"grad_norm": 1.9065883159637451,
|
| 107 |
+
"learning_rate": 0.00018426150121065375,
|
| 108 |
+
"loss": 2.8457,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.34,
|
| 113 |
+
"grad_norm": 2.126429796218872,
|
| 114 |
+
"learning_rate": 0.00018305084745762714,
|
| 115 |
+
"loss": 2.9638,
|
| 116 |
+
"step": 140
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.36,
|
| 120 |
+
"grad_norm": 1.9387011528015137,
|
| 121 |
+
"learning_rate": 0.00018184019370460052,
|
| 122 |
+
"loss": 2.7348,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.39,
|
| 127 |
+
"grad_norm": 1.923202633857727,
|
| 128 |
+
"learning_rate": 0.00018062953995157384,
|
| 129 |
+
"loss": 2.8489,
|
| 130 |
+
"step": 160
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.41,
|
| 134 |
+
"grad_norm": 2.581446409225464,
|
| 135 |
+
"learning_rate": 0.00017941888619854723,
|
| 136 |
+
"loss": 2.7744,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.44,
|
| 141 |
+
"grad_norm": 1.6987770795822144,
|
| 142 |
+
"learning_rate": 0.00017820823244552058,
|
| 143 |
+
"loss": 2.6428,
|
| 144 |
+
"step": 180
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.46,
|
| 148 |
+
"grad_norm": 1.9667104482650757,
|
| 149 |
+
"learning_rate": 0.00017699757869249396,
|
| 150 |
+
"loss": 2.6952,
|
| 151 |
+
"step": 190
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.48,
|
| 155 |
+
"grad_norm": 3.4282023906707764,
|
| 156 |
+
"learning_rate": 0.00017578692493946732,
|
| 157 |
+
"loss": 2.7372,
|
| 158 |
+
"step": 200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.48,
|
| 162 |
+
"eval_accuracy": 0.27615780445969124,
|
| 163 |
+
"eval_loss": 2.530054807662964,
|
| 164 |
+
"eval_runtime": 6.2338,
|
| 165 |
+
"eval_samples_per_second": 187.045,
|
| 166 |
+
"eval_steps_per_second": 23.421,
|
| 167 |
+
"step": 200
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.51,
|
| 171 |
+
"grad_norm": 1.9124583005905151,
|
| 172 |
+
"learning_rate": 0.0001745762711864407,
|
| 173 |
+
"loss": 2.6423,
|
| 174 |
+
"step": 210
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.53,
|
| 178 |
+
"grad_norm": 2.4269683361053467,
|
| 179 |
+
"learning_rate": 0.00017336561743341405,
|
| 180 |
+
"loss": 2.668,
|
| 181 |
+
"step": 220
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.56,
|
| 185 |
+
"grad_norm": 1.9838333129882812,
|
| 186 |
+
"learning_rate": 0.0001721549636803874,
|
| 187 |
+
"loss": 2.5786,
|
| 188 |
+
"step": 230
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.58,
|
| 192 |
+
"grad_norm": 3.200087070465088,
|
| 193 |
+
"learning_rate": 0.0001709443099273608,
|
| 194 |
+
"loss": 2.5938,
|
| 195 |
+
"step": 240
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.61,
|
| 199 |
+
"grad_norm": 2.93118953704834,
|
| 200 |
+
"learning_rate": 0.00016973365617433414,
|
| 201 |
+
"loss": 2.4526,
|
| 202 |
+
"step": 250
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.63,
|
| 206 |
+
"grad_norm": 2.555947780609131,
|
| 207 |
+
"learning_rate": 0.00016852300242130752,
|
| 208 |
+
"loss": 2.41,
|
| 209 |
+
"step": 260
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.65,
|
| 213 |
+
"grad_norm": 2.9446065425872803,
|
| 214 |
+
"learning_rate": 0.00016731234866828088,
|
| 215 |
+
"loss": 2.4537,
|
| 216 |
+
"step": 270
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.68,
|
| 220 |
+
"grad_norm": 3.393993377685547,
|
| 221 |
+
"learning_rate": 0.00016610169491525423,
|
| 222 |
+
"loss": 2.4256,
|
| 223 |
+
"step": 280
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 0.7,
|
| 227 |
+
"grad_norm": 2.721825122833252,
|
| 228 |
+
"learning_rate": 0.00016489104116222762,
|
| 229 |
+
"loss": 2.4719,
|
| 230 |
+
"step": 290
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.73,
|
| 234 |
+
"grad_norm": 3.2610368728637695,
|
| 235 |
+
"learning_rate": 0.00016368038740920097,
|
| 236 |
+
"loss": 2.4969,
|
| 237 |
+
"step": 300
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.73,
|
| 241 |
+
"eval_accuracy": 0.3439108061749571,
|
| 242 |
+
"eval_loss": 2.175961971282959,
|
| 243 |
+
"eval_runtime": 6.2327,
|
| 244 |
+
"eval_samples_per_second": 187.079,
|
| 245 |
+
"eval_steps_per_second": 23.425,
|
| 246 |
+
"step": 300
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.75,
|
| 250 |
+
"grad_norm": 3.067995309829712,
|
| 251 |
+
"learning_rate": 0.00016246973365617435,
|
| 252 |
+
"loss": 2.4904,
|
| 253 |
+
"step": 310
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.77,
|
| 257 |
+
"grad_norm": 2.7957141399383545,
|
| 258 |
+
"learning_rate": 0.0001612590799031477,
|
| 259 |
+
"loss": 2.3913,
|
| 260 |
+
"step": 320
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.8,
|
| 264 |
+
"grad_norm": 2.281586170196533,
|
| 265 |
+
"learning_rate": 0.0001600484261501211,
|
| 266 |
+
"loss": 2.1749,
|
| 267 |
+
"step": 330
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.82,
|
| 271 |
+
"grad_norm": 2.4833972454071045,
|
| 272 |
+
"learning_rate": 0.00015883777239709444,
|
| 273 |
+
"loss": 2.4058,
|
| 274 |
+
"step": 340
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.85,
|
| 278 |
+
"grad_norm": 2.5052073001861572,
|
| 279 |
+
"learning_rate": 0.0001576271186440678,
|
| 280 |
+
"loss": 2.3236,
|
| 281 |
+
"step": 350
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.87,
|
| 285 |
+
"grad_norm": 2.479684352874756,
|
| 286 |
+
"learning_rate": 0.00015641646489104115,
|
| 287 |
+
"loss": 2.373,
|
| 288 |
+
"step": 360
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.9,
|
| 292 |
+
"grad_norm": 3.6352992057800293,
|
| 293 |
+
"learning_rate": 0.00015520581113801453,
|
| 294 |
+
"loss": 2.3282,
|
| 295 |
+
"step": 370
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.92,
|
| 299 |
+
"grad_norm": 2.748934030532837,
|
| 300 |
+
"learning_rate": 0.00015399515738498791,
|
| 301 |
+
"loss": 2.2062,
|
| 302 |
+
"step": 380
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.94,
|
| 306 |
+
"grad_norm": 2.0645978450775146,
|
| 307 |
+
"learning_rate": 0.00015278450363196127,
|
| 308 |
+
"loss": 2.1253,
|
| 309 |
+
"step": 390
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.97,
|
| 313 |
+
"grad_norm": 2.2856009006500244,
|
| 314 |
+
"learning_rate": 0.00015157384987893465,
|
| 315 |
+
"loss": 2.1973,
|
| 316 |
+
"step": 400
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.97,
|
| 320 |
+
"eval_accuracy": 0.37564322469982847,
|
| 321 |
+
"eval_loss": 2.0102577209472656,
|
| 322 |
+
"eval_runtime": 5.9741,
|
| 323 |
+
"eval_samples_per_second": 195.175,
|
| 324 |
+
"eval_steps_per_second": 24.439,
|
| 325 |
+
"step": 400
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.99,
|
| 329 |
+
"grad_norm": 2.866960048675537,
|
| 330 |
+
"learning_rate": 0.00015036319612590798,
|
| 331 |
+
"loss": 2.214,
|
| 332 |
+
"step": 410
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.02,
|
| 336 |
+
"grad_norm": 3.171844482421875,
|
| 337 |
+
"learning_rate": 0.00014915254237288136,
|
| 338 |
+
"loss": 2.0948,
|
| 339 |
+
"step": 420
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.04,
|
| 343 |
+
"grad_norm": 3.6916253566741943,
|
| 344 |
+
"learning_rate": 0.00014794188861985471,
|
| 345 |
+
"loss": 2.0649,
|
| 346 |
+
"step": 430
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.07,
|
| 350 |
+
"grad_norm": 2.3281314373016357,
|
| 351 |
+
"learning_rate": 0.0001467312348668281,
|
| 352 |
+
"loss": 2.0633,
|
| 353 |
+
"step": 440
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.09,
|
| 357 |
+
"grad_norm": 3.370180368423462,
|
| 358 |
+
"learning_rate": 0.00014552058111380148,
|
| 359 |
+
"loss": 1.9949,
|
| 360 |
+
"step": 450
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.11,
|
| 364 |
+
"grad_norm": 2.5389626026153564,
|
| 365 |
+
"learning_rate": 0.00014430992736077483,
|
| 366 |
+
"loss": 2.086,
|
| 367 |
+
"step": 460
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.14,
|
| 371 |
+
"grad_norm": 2.47526216506958,
|
| 372 |
+
"learning_rate": 0.00014309927360774819,
|
| 373 |
+
"loss": 2.0443,
|
| 374 |
+
"step": 470
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.16,
|
| 378 |
+
"grad_norm": 2.821577548980713,
|
| 379 |
+
"learning_rate": 0.00014188861985472154,
|
| 380 |
+
"loss": 2.0808,
|
| 381 |
+
"step": 480
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.19,
|
| 385 |
+
"grad_norm": 2.978994369506836,
|
| 386 |
+
"learning_rate": 0.00014067796610169492,
|
| 387 |
+
"loss": 2.1278,
|
| 388 |
+
"step": 490
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.21,
|
| 392 |
+
"grad_norm": 3.1431379318237305,
|
| 393 |
+
"learning_rate": 0.00013946731234866828,
|
| 394 |
+
"loss": 1.8847,
|
| 395 |
+
"step": 500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.21,
|
| 399 |
+
"eval_accuracy": 0.41080617495711835,
|
| 400 |
+
"eval_loss": 1.8402307033538818,
|
| 401 |
+
"eval_runtime": 6.2119,
|
| 402 |
+
"eval_samples_per_second": 187.705,
|
| 403 |
+
"eval_steps_per_second": 23.503,
|
| 404 |
+
"step": 500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.23,
|
| 408 |
+
"grad_norm": 3.1350502967834473,
|
| 409 |
+
"learning_rate": 0.00013825665859564166,
|
| 410 |
+
"loss": 2.02,
|
| 411 |
+
"step": 510
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 1.26,
|
| 415 |
+
"grad_norm": 2.63952374458313,
|
| 416 |
+
"learning_rate": 0.00013704600484261504,
|
| 417 |
+
"loss": 2.1684,
|
| 418 |
+
"step": 520
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 1.28,
|
| 422 |
+
"grad_norm": 2.7914199829101562,
|
| 423 |
+
"learning_rate": 0.00013583535108958837,
|
| 424 |
+
"loss": 1.8532,
|
| 425 |
+
"step": 530
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 1.31,
|
| 429 |
+
"grad_norm": 4.124698638916016,
|
| 430 |
+
"learning_rate": 0.00013462469733656175,
|
| 431 |
+
"loss": 1.9593,
|
| 432 |
+
"step": 540
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 1.33,
|
| 436 |
+
"grad_norm": 3.0953214168548584,
|
| 437 |
+
"learning_rate": 0.0001334140435835351,
|
| 438 |
+
"loss": 2.0143,
|
| 439 |
+
"step": 550
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 1.36,
|
| 443 |
+
"grad_norm": 3.626241683959961,
|
| 444 |
+
"learning_rate": 0.00013220338983050849,
|
| 445 |
+
"loss": 2.0349,
|
| 446 |
+
"step": 560
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 1.38,
|
| 450 |
+
"grad_norm": 3.22306752204895,
|
| 451 |
+
"learning_rate": 0.00013099273607748184,
|
| 452 |
+
"loss": 1.9283,
|
| 453 |
+
"step": 570
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 1.4,
|
| 457 |
+
"grad_norm": 2.6860299110412598,
|
| 458 |
+
"learning_rate": 0.00012978208232445522,
|
| 459 |
+
"loss": 1.9022,
|
| 460 |
+
"step": 580
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 1.43,
|
| 464 |
+
"grad_norm": 3.2099533081054688,
|
| 465 |
+
"learning_rate": 0.00012857142857142858,
|
| 466 |
+
"loss": 1.9102,
|
| 467 |
+
"step": 590
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 1.45,
|
| 471 |
+
"grad_norm": 2.5889129638671875,
|
| 472 |
+
"learning_rate": 0.00012736077481840193,
|
| 473 |
+
"loss": 1.746,
|
| 474 |
+
"step": 600
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.45,
|
| 478 |
+
"eval_accuracy": 0.48027444253859347,
|
| 479 |
+
"eval_loss": 1.7051318883895874,
|
| 480 |
+
"eval_runtime": 6.0914,
|
| 481 |
+
"eval_samples_per_second": 191.418,
|
| 482 |
+
"eval_steps_per_second": 23.968,
|
| 483 |
+
"step": 600
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 1.48,
|
| 487 |
+
"grad_norm": 2.6496353149414062,
|
| 488 |
+
"learning_rate": 0.0001261501210653753,
|
| 489 |
+
"loss": 1.8087,
|
| 490 |
+
"step": 610
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 1.5,
|
| 494 |
+
"grad_norm": 2.2695322036743164,
|
| 495 |
+
"learning_rate": 0.00012493946731234867,
|
| 496 |
+
"loss": 1.9172,
|
| 497 |
+
"step": 620
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.53,
|
| 501 |
+
"grad_norm": 3.144073724746704,
|
| 502 |
+
"learning_rate": 0.00012372881355932205,
|
| 503 |
+
"loss": 1.8943,
|
| 504 |
+
"step": 630
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.55,
|
| 508 |
+
"grad_norm": 2.9001333713531494,
|
| 509 |
+
"learning_rate": 0.0001225181598062954,
|
| 510 |
+
"loss": 1.9463,
|
| 511 |
+
"step": 640
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.57,
|
| 515 |
+
"grad_norm": 2.5096278190612793,
|
| 516 |
+
"learning_rate": 0.00012130750605326877,
|
| 517 |
+
"loss": 1.8045,
|
| 518 |
+
"step": 650
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.6,
|
| 522 |
+
"grad_norm": 2.2238059043884277,
|
| 523 |
+
"learning_rate": 0.00012009685230024215,
|
| 524 |
+
"loss": 1.9322,
|
| 525 |
+
"step": 660
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 1.62,
|
| 529 |
+
"grad_norm": 2.7545368671417236,
|
| 530 |
+
"learning_rate": 0.00011888619854721549,
|
| 531 |
+
"loss": 1.7305,
|
| 532 |
+
"step": 670
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 1.65,
|
| 536 |
+
"grad_norm": 2.8309366703033447,
|
| 537 |
+
"learning_rate": 0.00011767554479418887,
|
| 538 |
+
"loss": 1.8587,
|
| 539 |
+
"step": 680
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 1.67,
|
| 543 |
+
"grad_norm": 5.093832492828369,
|
| 544 |
+
"learning_rate": 0.00011646489104116223,
|
| 545 |
+
"loss": 1.8362,
|
| 546 |
+
"step": 690
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 1.69,
|
| 550 |
+
"grad_norm": 2.4374847412109375,
|
| 551 |
+
"learning_rate": 0.0001152542372881356,
|
| 552 |
+
"loss": 1.8698,
|
| 553 |
+
"step": 700
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.69,
|
| 557 |
+
"eval_accuracy": 0.4888507718696398,
|
| 558 |
+
"eval_loss": 1.5985045433044434,
|
| 559 |
+
"eval_runtime": 6.4332,
|
| 560 |
+
"eval_samples_per_second": 181.249,
|
| 561 |
+
"eval_steps_per_second": 22.695,
|
| 562 |
+
"step": 700
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.72,
|
| 566 |
+
"grad_norm": 2.8519837856292725,
|
| 567 |
+
"learning_rate": 0.00011404358353510895,
|
| 568 |
+
"loss": 1.8736,
|
| 569 |
+
"step": 710
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.74,
|
| 573 |
+
"grad_norm": 2.8379719257354736,
|
| 574 |
+
"learning_rate": 0.00011283292978208233,
|
| 575 |
+
"loss": 1.6395,
|
| 576 |
+
"step": 720
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.77,
|
| 580 |
+
"grad_norm": 3.884648323059082,
|
| 581 |
+
"learning_rate": 0.00011174334140435836,
|
| 582 |
+
"loss": 1.7938,
|
| 583 |
+
"step": 730
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.79,
|
| 587 |
+
"grad_norm": 3.2592883110046387,
|
| 588 |
+
"learning_rate": 0.00011053268765133173,
|
| 589 |
+
"loss": 1.6813,
|
| 590 |
+
"step": 740
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.82,
|
| 594 |
+
"grad_norm": 5.118261337280273,
|
| 595 |
+
"learning_rate": 0.00010932203389830508,
|
| 596 |
+
"loss": 1.9414,
|
| 597 |
+
"step": 750
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.84,
|
| 601 |
+
"grad_norm": 2.822026491165161,
|
| 602 |
+
"learning_rate": 0.00010811138014527846,
|
| 603 |
+
"loss": 1.7598,
|
| 604 |
+
"step": 760
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.86,
|
| 608 |
+
"grad_norm": 2.8540070056915283,
|
| 609 |
+
"learning_rate": 0.00010690072639225182,
|
| 610 |
+
"loss": 1.7024,
|
| 611 |
+
"step": 770
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.89,
|
| 615 |
+
"grad_norm": 4.354470252990723,
|
| 616 |
+
"learning_rate": 0.00010569007263922519,
|
| 617 |
+
"loss": 1.8987,
|
| 618 |
+
"step": 780
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.91,
|
| 622 |
+
"grad_norm": 3.528857707977295,
|
| 623 |
+
"learning_rate": 0.00010447941888619854,
|
| 624 |
+
"loss": 1.7933,
|
| 625 |
+
"step": 790
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.94,
|
| 629 |
+
"grad_norm": 2.76985764503479,
|
| 630 |
+
"learning_rate": 0.00010326876513317192,
|
| 631 |
+
"loss": 1.7261,
|
| 632 |
+
"step": 800
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.94,
|
| 636 |
+
"eval_accuracy": 0.5840480274442539,
|
| 637 |
+
"eval_loss": 1.4311938285827637,
|
| 638 |
+
"eval_runtime": 6.2955,
|
| 639 |
+
"eval_samples_per_second": 185.213,
|
| 640 |
+
"eval_steps_per_second": 23.191,
|
| 641 |
+
"step": 800
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 1.96,
|
| 645 |
+
"grad_norm": 3.15104079246521,
|
| 646 |
+
"learning_rate": 0.00010205811138014529,
|
| 647 |
+
"loss": 1.8079,
|
| 648 |
+
"step": 810
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 1.99,
|
| 652 |
+
"grad_norm": 3.0211942195892334,
|
| 653 |
+
"learning_rate": 0.00010084745762711865,
|
| 654 |
+
"loss": 1.611,
|
| 655 |
+
"step": 820
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.01,
|
| 659 |
+
"grad_norm": 2.527198076248169,
|
| 660 |
+
"learning_rate": 9.963680387409201e-05,
|
| 661 |
+
"loss": 1.7344,
|
| 662 |
+
"step": 830
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.03,
|
| 666 |
+
"grad_norm": 3.654705762863159,
|
| 667 |
+
"learning_rate": 9.842615012106537e-05,
|
| 668 |
+
"loss": 1.5921,
|
| 669 |
+
"step": 840
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.06,
|
| 673 |
+
"grad_norm": 2.6901042461395264,
|
| 674 |
+
"learning_rate": 9.721549636803875e-05,
|
| 675 |
+
"loss": 1.5688,
|
| 676 |
+
"step": 850
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 2.08,
|
| 680 |
+
"grad_norm": 2.830200672149658,
|
| 681 |
+
"learning_rate": 9.600484261501212e-05,
|
| 682 |
+
"loss": 1.5546,
|
| 683 |
+
"step": 860
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 2.11,
|
| 687 |
+
"grad_norm": 3.2287344932556152,
|
| 688 |
+
"learning_rate": 9.479418886198547e-05,
|
| 689 |
+
"loss": 1.5714,
|
| 690 |
+
"step": 870
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 2.13,
|
| 694 |
+
"grad_norm": 3.661449432373047,
|
| 695 |
+
"learning_rate": 9.358353510895884e-05,
|
| 696 |
+
"loss": 1.4973,
|
| 697 |
+
"step": 880
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 2.15,
|
| 701 |
+
"grad_norm": 6.353243827819824,
|
| 702 |
+
"learning_rate": 9.237288135593221e-05,
|
| 703 |
+
"loss": 1.5294,
|
| 704 |
+
"step": 890
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 2.18,
|
| 708 |
+
"grad_norm": 3.703733444213867,
|
| 709 |
+
"learning_rate": 9.116222760290558e-05,
|
| 710 |
+
"loss": 1.7385,
|
| 711 |
+
"step": 900
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 2.18,
|
| 715 |
+
"eval_accuracy": 0.6286449399656947,
|
| 716 |
+
"eval_loss": 1.3585376739501953,
|
| 717 |
+
"eval_runtime": 5.9781,
|
| 718 |
+
"eval_samples_per_second": 195.046,
|
| 719 |
+
"eval_steps_per_second": 24.423,
|
| 720 |
+
"step": 900
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 2.2,
|
| 724 |
+
"grad_norm": 2.730365514755249,
|
| 725 |
+
"learning_rate": 8.995157384987893e-05,
|
| 726 |
+
"loss": 1.626,
|
| 727 |
+
"step": 910
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 2.23,
|
| 731 |
+
"grad_norm": 4.335669040679932,
|
| 732 |
+
"learning_rate": 8.874092009685231e-05,
|
| 733 |
+
"loss": 1.5823,
|
| 734 |
+
"step": 920
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 2.25,
|
| 738 |
+
"grad_norm": 2.272915840148926,
|
| 739 |
+
"learning_rate": 8.753026634382567e-05,
|
| 740 |
+
"loss": 1.47,
|
| 741 |
+
"step": 930
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 2.28,
|
| 745 |
+
"grad_norm": 3.335453510284424,
|
| 746 |
+
"learning_rate": 8.631961259079904e-05,
|
| 747 |
+
"loss": 1.4733,
|
| 748 |
+
"step": 940
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 2.3,
|
| 752 |
+
"grad_norm": 5.18184232711792,
|
| 753 |
+
"learning_rate": 8.51089588377724e-05,
|
| 754 |
+
"loss": 1.3798,
|
| 755 |
+
"step": 950
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 2.32,
|
| 759 |
+
"grad_norm": 3.79761004447937,
|
| 760 |
+
"learning_rate": 8.389830508474577e-05,
|
| 761 |
+
"loss": 1.5103,
|
| 762 |
+
"step": 960
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 2.35,
|
| 766 |
+
"grad_norm": 2.568056344985962,
|
| 767 |
+
"learning_rate": 8.268765133171913e-05,
|
| 768 |
+
"loss": 1.5016,
|
| 769 |
+
"step": 970
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 2.37,
|
| 773 |
+
"grad_norm": 4.231459140777588,
|
| 774 |
+
"learning_rate": 8.14769975786925e-05,
|
| 775 |
+
"loss": 1.4617,
|
| 776 |
+
"step": 980
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 2.4,
|
| 780 |
+
"grad_norm": 3.2914044857025146,
|
| 781 |
+
"learning_rate": 8.026634382566586e-05,
|
| 782 |
+
"loss": 1.5527,
|
| 783 |
+
"step": 990
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 2.42,
|
| 787 |
+
"grad_norm": 2.967702627182007,
|
| 788 |
+
"learning_rate": 7.905569007263923e-05,
|
| 789 |
+
"loss": 1.5873,
|
| 790 |
+
"step": 1000
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 2.42,
|
| 794 |
+
"eval_accuracy": 0.6758147512864494,
|
| 795 |
+
"eval_loss": 1.2374264001846313,
|
| 796 |
+
"eval_runtime": 6.2974,
|
| 797 |
+
"eval_samples_per_second": 185.155,
|
| 798 |
+
"eval_steps_per_second": 23.184,
|
| 799 |
+
"step": 1000
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 2.45,
|
| 803 |
+
"grad_norm": 2.7834739685058594,
|
| 804 |
+
"learning_rate": 7.78450363196126e-05,
|
| 805 |
+
"loss": 1.4255,
|
| 806 |
+
"step": 1010
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.47,
|
| 810 |
+
"grad_norm": 3.380810260772705,
|
| 811 |
+
"learning_rate": 7.663438256658597e-05,
|
| 812 |
+
"loss": 1.4528,
|
| 813 |
+
"step": 1020
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 2.49,
|
| 817 |
+
"grad_norm": 3.3973748683929443,
|
| 818 |
+
"learning_rate": 7.542372881355932e-05,
|
| 819 |
+
"loss": 1.5726,
|
| 820 |
+
"step": 1030
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 2.52,
|
| 824 |
+
"grad_norm": 2.9069502353668213,
|
| 825 |
+
"learning_rate": 7.421307506053269e-05,
|
| 826 |
+
"loss": 1.2987,
|
| 827 |
+
"step": 1040
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 2.54,
|
| 831 |
+
"grad_norm": 2.8832297325134277,
|
| 832 |
+
"learning_rate": 7.300242130750606e-05,
|
| 833 |
+
"loss": 1.437,
|
| 834 |
+
"step": 1050
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 2.57,
|
| 838 |
+
"grad_norm": 3.137310743331909,
|
| 839 |
+
"learning_rate": 7.179176755447942e-05,
|
| 840 |
+
"loss": 1.5,
|
| 841 |
+
"step": 1060
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 2.59,
|
| 845 |
+
"grad_norm": 3.156430244445801,
|
| 846 |
+
"learning_rate": 7.058111380145279e-05,
|
| 847 |
+
"loss": 1.341,
|
| 848 |
+
"step": 1070
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 2.62,
|
| 852 |
+
"grad_norm": 3.470303535461426,
|
| 853 |
+
"learning_rate": 6.937046004842616e-05,
|
| 854 |
+
"loss": 1.3986,
|
| 855 |
+
"step": 1080
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 2.64,
|
| 859 |
+
"grad_norm": 3.426010847091675,
|
| 860 |
+
"learning_rate": 6.815980629539952e-05,
|
| 861 |
+
"loss": 1.3874,
|
| 862 |
+
"step": 1090
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 2.66,
|
| 866 |
+
"grad_norm": 3.8181042671203613,
|
| 867 |
+
"learning_rate": 6.694915254237288e-05,
|
| 868 |
+
"loss": 1.4775,
|
| 869 |
+
"step": 1100
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 2.66,
|
| 873 |
+
"eval_accuracy": 0.7024013722126929,
|
| 874 |
+
"eval_loss": 1.1351556777954102,
|
| 875 |
+
"eval_runtime": 6.2887,
|
| 876 |
+
"eval_samples_per_second": 185.412,
|
| 877 |
+
"eval_steps_per_second": 23.216,
|
| 878 |
+
"step": 1100
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 2.69,
|
| 882 |
+
"grad_norm": 3.4228086471557617,
|
| 883 |
+
"learning_rate": 6.573849878934625e-05,
|
| 884 |
+
"loss": 1.4804,
|
| 885 |
+
"step": 1110
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 2.71,
|
| 889 |
+
"grad_norm": 4.945833206176758,
|
| 890 |
+
"learning_rate": 6.45278450363196e-05,
|
| 891 |
+
"loss": 1.2617,
|
| 892 |
+
"step": 1120
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 2.74,
|
| 896 |
+
"grad_norm": 2.712095022201538,
|
| 897 |
+
"learning_rate": 6.331719128329297e-05,
|
| 898 |
+
"loss": 1.4254,
|
| 899 |
+
"step": 1130
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 2.76,
|
| 903 |
+
"grad_norm": 3.2312748432159424,
|
| 904 |
+
"learning_rate": 6.210653753026636e-05,
|
| 905 |
+
"loss": 1.4141,
|
| 906 |
+
"step": 1140
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 2.78,
|
| 910 |
+
"grad_norm": 2.4630300998687744,
|
| 911 |
+
"learning_rate": 6.089588377723972e-05,
|
| 912 |
+
"loss": 1.3438,
|
| 913 |
+
"step": 1150
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 2.81,
|
| 917 |
+
"grad_norm": 2.9009976387023926,
|
| 918 |
+
"learning_rate": 5.968523002421308e-05,
|
| 919 |
+
"loss": 1.3625,
|
| 920 |
+
"step": 1160
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 2.83,
|
| 924 |
+
"grad_norm": 5.364362716674805,
|
| 925 |
+
"learning_rate": 5.8474576271186446e-05,
|
| 926 |
+
"loss": 1.4056,
|
| 927 |
+
"step": 1170
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 2.86,
|
| 931 |
+
"grad_norm": 3.0310747623443604,
|
| 932 |
+
"learning_rate": 5.726392251815981e-05,
|
| 933 |
+
"loss": 1.2943,
|
| 934 |
+
"step": 1180
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 2.88,
|
| 938 |
+
"grad_norm": 2.7472984790802,
|
| 939 |
+
"learning_rate": 5.605326876513317e-05,
|
| 940 |
+
"loss": 1.4934,
|
| 941 |
+
"step": 1190
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 2.91,
|
| 945 |
+
"grad_norm": 2.9528918266296387,
|
| 946 |
+
"learning_rate": 5.484261501210654e-05,
|
| 947 |
+
"loss": 1.2697,
|
| 948 |
+
"step": 1200
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 2.91,
|
| 952 |
+
"eval_accuracy": 0.70926243567753,
|
| 953 |
+
"eval_loss": 1.104396104812622,
|
| 954 |
+
"eval_runtime": 6.0071,
|
| 955 |
+
"eval_samples_per_second": 194.105,
|
| 956 |
+
"eval_steps_per_second": 24.305,
|
| 957 |
+
"step": 1200
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 2.93,
|
| 961 |
+
"grad_norm": 2.5816805362701416,
|
| 962 |
+
"learning_rate": 5.363196125907991e-05,
|
| 963 |
+
"loss": 1.3362,
|
| 964 |
+
"step": 1210
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 2.95,
|
| 968 |
+
"grad_norm": 3.5116188526153564,
|
| 969 |
+
"learning_rate": 5.242130750605327e-05,
|
| 970 |
+
"loss": 1.3128,
|
| 971 |
+
"step": 1220
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 2.98,
|
| 975 |
+
"grad_norm": 2.873042583465576,
|
| 976 |
+
"learning_rate": 5.121065375302664e-05,
|
| 977 |
+
"loss": 1.3257,
|
| 978 |
+
"step": 1230
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 3.0,
|
| 982 |
+
"grad_norm": 6.232132434844971,
|
| 983 |
+
"learning_rate": 5e-05,
|
| 984 |
+
"loss": 1.256,
|
| 985 |
+
"step": 1240
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 3.03,
|
| 989 |
+
"grad_norm": 2.3054957389831543,
|
| 990 |
+
"learning_rate": 4.8789346246973364e-05,
|
| 991 |
+
"loss": 1.1805,
|
| 992 |
+
"step": 1250
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 3.05,
|
| 996 |
+
"grad_norm": 3.0687952041625977,
|
| 997 |
+
"learning_rate": 4.757869249394674e-05,
|
| 998 |
+
"loss": 1.0767,
|
| 999 |
+
"step": 1260
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 3.08,
|
| 1003 |
+
"grad_norm": 3.774822235107422,
|
| 1004 |
+
"learning_rate": 4.63680387409201e-05,
|
| 1005 |
+
"loss": 1.311,
|
| 1006 |
+
"step": 1270
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 3.1,
|
| 1010 |
+
"grad_norm": 4.785544395446777,
|
| 1011 |
+
"learning_rate": 4.515738498789346e-05,
|
| 1012 |
+
"loss": 1.2997,
|
| 1013 |
+
"step": 1280
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 3.12,
|
| 1017 |
+
"grad_norm": 3.4525294303894043,
|
| 1018 |
+
"learning_rate": 4.394673123486683e-05,
|
| 1019 |
+
"loss": 1.2039,
|
| 1020 |
+
"step": 1290
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 3.15,
|
| 1024 |
+
"grad_norm": 3.312502861022949,
|
| 1025 |
+
"learning_rate": 4.27360774818402e-05,
|
| 1026 |
+
"loss": 1.2137,
|
| 1027 |
+
"step": 1300
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 3.15,
|
| 1031 |
+
"eval_accuracy": 0.7615780445969125,
|
| 1032 |
+
"eval_loss": 1.0005759000778198,
|
| 1033 |
+
"eval_runtime": 6.3563,
|
| 1034 |
+
"eval_samples_per_second": 183.44,
|
| 1035 |
+
"eval_steps_per_second": 22.969,
|
| 1036 |
+
"step": 1300
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 3.17,
|
| 1040 |
+
"grad_norm": 3.375433921813965,
|
| 1041 |
+
"learning_rate": 4.152542372881356e-05,
|
| 1042 |
+
"loss": 1.2714,
|
| 1043 |
+
"step": 1310
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 3.2,
|
| 1047 |
+
"grad_norm": 2.5909006595611572,
|
| 1048 |
+
"learning_rate": 4.0314769975786926e-05,
|
| 1049 |
+
"loss": 1.3154,
|
| 1050 |
+
"step": 1320
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 3.22,
|
| 1054 |
+
"grad_norm": 3.0990185737609863,
|
| 1055 |
+
"learning_rate": 3.910411622276029e-05,
|
| 1056 |
+
"loss": 1.144,
|
| 1057 |
+
"step": 1330
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 3.24,
|
| 1061 |
+
"grad_norm": 1.911260962486267,
|
| 1062 |
+
"learning_rate": 3.789346246973366e-05,
|
| 1063 |
+
"loss": 1.0008,
|
| 1064 |
+
"step": 1340
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 3.27,
|
| 1068 |
+
"grad_norm": 2.93192458152771,
|
| 1069 |
+
"learning_rate": 3.6682808716707024e-05,
|
| 1070 |
+
"loss": 1.0603,
|
| 1071 |
+
"step": 1350
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 3.29,
|
| 1075 |
+
"grad_norm": 3.3576924800872803,
|
| 1076 |
+
"learning_rate": 3.5472154963680385e-05,
|
| 1077 |
+
"loss": 1.2791,
|
| 1078 |
+
"step": 1360
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 3.32,
|
| 1082 |
+
"grad_norm": 2.8567686080932617,
|
| 1083 |
+
"learning_rate": 3.426150121065376e-05,
|
| 1084 |
+
"loss": 1.1223,
|
| 1085 |
+
"step": 1370
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 3.34,
|
| 1089 |
+
"grad_norm": 2.735358953475952,
|
| 1090 |
+
"learning_rate": 3.305084745762712e-05,
|
| 1091 |
+
"loss": 1.2043,
|
| 1092 |
+
"step": 1380
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 3.37,
|
| 1096 |
+
"grad_norm": 3.374582529067993,
|
| 1097 |
+
"learning_rate": 3.184019370460048e-05,
|
| 1098 |
+
"loss": 1.0495,
|
| 1099 |
+
"step": 1390
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 3.39,
|
| 1103 |
+
"grad_norm": 4.9084792137146,
|
| 1104 |
+
"learning_rate": 3.062953995157385e-05,
|
| 1105 |
+
"loss": 1.423,
|
| 1106 |
+
"step": 1400
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 3.39,
|
| 1110 |
+
"eval_accuracy": 0.774442538593482,
|
| 1111 |
+
"eval_loss": 0.9588848352432251,
|
| 1112 |
+
"eval_runtime": 6.2494,
|
| 1113 |
+
"eval_samples_per_second": 186.579,
|
| 1114 |
+
"eval_steps_per_second": 23.362,
|
| 1115 |
+
"step": 1400
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 3.41,
|
| 1119 |
+
"grad_norm": 4.47416353225708,
|
| 1120 |
+
"learning_rate": 2.941888619854722e-05,
|
| 1121 |
+
"loss": 1.2965,
|
| 1122 |
+
"step": 1410
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 3.44,
|
| 1126 |
+
"grad_norm": 2.692729949951172,
|
| 1127 |
+
"learning_rate": 2.8208232445520583e-05,
|
| 1128 |
+
"loss": 1.1812,
|
| 1129 |
+
"step": 1420
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 3.46,
|
| 1133 |
+
"grad_norm": 3.5278244018554688,
|
| 1134 |
+
"learning_rate": 2.6997578692493948e-05,
|
| 1135 |
+
"loss": 1.2515,
|
| 1136 |
+
"step": 1430
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 3.49,
|
| 1140 |
+
"grad_norm": 2.9056203365325928,
|
| 1141 |
+
"learning_rate": 2.5786924939467316e-05,
|
| 1142 |
+
"loss": 1.0617,
|
| 1143 |
+
"step": 1440
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 3.51,
|
| 1147 |
+
"grad_norm": 2.6366896629333496,
|
| 1148 |
+
"learning_rate": 2.457627118644068e-05,
|
| 1149 |
+
"loss": 1.0449,
|
| 1150 |
+
"step": 1450
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 3.54,
|
| 1154 |
+
"grad_norm": 3.593003034591675,
|
| 1155 |
+
"learning_rate": 2.3365617433414045e-05,
|
| 1156 |
+
"loss": 1.1273,
|
| 1157 |
+
"step": 1460
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 3.56,
|
| 1161 |
+
"grad_norm": 3.5506863594055176,
|
| 1162 |
+
"learning_rate": 2.215496368038741e-05,
|
| 1163 |
+
"loss": 1.2625,
|
| 1164 |
+
"step": 1470
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 3.58,
|
| 1168 |
+
"grad_norm": 4.686192989349365,
|
| 1169 |
+
"learning_rate": 2.0944309927360775e-05,
|
| 1170 |
+
"loss": 1.1439,
|
| 1171 |
+
"step": 1480
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 3.61,
|
| 1175 |
+
"grad_norm": 3.072838068008423,
|
| 1176 |
+
"learning_rate": 1.9733656174334143e-05,
|
| 1177 |
+
"loss": 1.2008,
|
| 1178 |
+
"step": 1490
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 3.63,
|
| 1182 |
+
"grad_norm": 4.130647659301758,
|
| 1183 |
+
"learning_rate": 1.8523002421307507e-05,
|
| 1184 |
+
"loss": 1.0098,
|
| 1185 |
+
"step": 1500
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 3.63,
|
| 1189 |
+
"eval_accuracy": 0.7684391080617495,
|
| 1190 |
+
"eval_loss": 0.9360153675079346,
|
| 1191 |
+
"eval_runtime": 5.9954,
|
| 1192 |
+
"eval_samples_per_second": 194.481,
|
| 1193 |
+
"eval_steps_per_second": 24.352,
|
| 1194 |
+
"step": 1500
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 3.66,
|
| 1198 |
+
"grad_norm": 2.432633638381958,
|
| 1199 |
+
"learning_rate": 1.7312348668280872e-05,
|
| 1200 |
+
"loss": 1.0802,
|
| 1201 |
+
"step": 1510
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 3.68,
|
| 1205 |
+
"grad_norm": 3.6661131381988525,
|
| 1206 |
+
"learning_rate": 1.6101694915254237e-05,
|
| 1207 |
+
"loss": 1.0655,
|
| 1208 |
+
"step": 1520
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 3.7,
|
| 1212 |
+
"grad_norm": 3.967733860015869,
|
| 1213 |
+
"learning_rate": 1.4891041162227603e-05,
|
| 1214 |
+
"loss": 1.1482,
|
| 1215 |
+
"step": 1530
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 3.73,
|
| 1219 |
+
"grad_norm": 3.776456832885742,
|
| 1220 |
+
"learning_rate": 1.3680387409200971e-05,
|
| 1221 |
+
"loss": 1.2236,
|
| 1222 |
+
"step": 1540
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 3.75,
|
| 1226 |
+
"grad_norm": 3.1570096015930176,
|
| 1227 |
+
"learning_rate": 1.2469733656174334e-05,
|
| 1228 |
+
"loss": 1.0433,
|
| 1229 |
+
"step": 1550
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 3.78,
|
| 1233 |
+
"grad_norm": 3.3112399578094482,
|
| 1234 |
+
"learning_rate": 1.12590799031477e-05,
|
| 1235 |
+
"loss": 1.1766,
|
| 1236 |
+
"step": 1560
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 3.8,
|
| 1240 |
+
"grad_norm": 3.405649185180664,
|
| 1241 |
+
"learning_rate": 1.0048426150121065e-05,
|
| 1242 |
+
"loss": 1.1755,
|
| 1243 |
+
"step": 1570
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 3.83,
|
| 1247 |
+
"grad_norm": 2.6833651065826416,
|
| 1248 |
+
"learning_rate": 8.837772397094432e-06,
|
| 1249 |
+
"loss": 1.0593,
|
| 1250 |
+
"step": 1580
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 3.85,
|
| 1254 |
+
"grad_norm": 3.3236443996429443,
|
| 1255 |
+
"learning_rate": 7.627118644067798e-06,
|
| 1256 |
+
"loss": 1.1001,
|
| 1257 |
+
"step": 1590
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 3.87,
|
| 1261 |
+
"grad_norm": 3.5733933448791504,
|
| 1262 |
+
"learning_rate": 6.4164648910411625e-06,
|
| 1263 |
+
"loss": 1.1325,
|
| 1264 |
+
"step": 1600
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 3.87,
|
| 1268 |
+
"eval_accuracy": 0.7710120068610634,
|
| 1269 |
+
"eval_loss": 0.912144124507904,
|
| 1270 |
+
"eval_runtime": 6.3188,
|
| 1271 |
+
"eval_samples_per_second": 184.528,
|
| 1272 |
+
"eval_steps_per_second": 23.106,
|
| 1273 |
+
"step": 1600
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 3.9,
|
| 1277 |
+
"grad_norm": 3.3235766887664795,
|
| 1278 |
+
"learning_rate": 5.205811138014528e-06,
|
| 1279 |
+
"loss": 1.1434,
|
| 1280 |
+
"step": 1610
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 3.92,
|
| 1284 |
+
"grad_norm": 5.47670841217041,
|
| 1285 |
+
"learning_rate": 3.9951573849878936e-06,
|
| 1286 |
+
"loss": 1.0415,
|
| 1287 |
+
"step": 1620
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 3.95,
|
| 1291 |
+
"grad_norm": 2.83181095123291,
|
| 1292 |
+
"learning_rate": 2.784503631961259e-06,
|
| 1293 |
+
"loss": 1.0888,
|
| 1294 |
+
"step": 1630
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 3.97,
|
| 1298 |
+
"grad_norm": 4.57571268081665,
|
| 1299 |
+
"learning_rate": 1.5738498789346248e-06,
|
| 1300 |
+
"loss": 1.0908,
|
| 1301 |
+
"step": 1640
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 4.0,
|
| 1305 |
+
"grad_norm": 3.1416895389556885,
|
| 1306 |
+
"learning_rate": 3.6319612590799036e-07,
|
| 1307 |
+
"loss": 0.9855,
|
| 1308 |
+
"step": 1650
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 4.0,
|
| 1312 |
+
"step": 1652,
|
| 1313 |
+
"total_flos": 2.047635634195759e+18,
|
| 1314 |
+
"train_loss": 1.8191430680543978,
|
| 1315 |
+
"train_runtime": 594.0822,
|
| 1316 |
+
"train_samples_per_second": 44.458,
|
| 1317 |
+
"train_steps_per_second": 2.781
|
| 1318 |
+
}
|
| 1319 |
+
],
|
| 1320 |
+
"logging_steps": 10,
|
| 1321 |
+
"max_steps": 1652,
|
| 1322 |
+
"num_input_tokens_seen": 0,
|
| 1323 |
+
"num_train_epochs": 4,
|
| 1324 |
+
"save_steps": 100,
|
| 1325 |
+
"total_flos": 2.047635634195759e+18,
|
| 1326 |
+
"train_batch_size": 16,
|
| 1327 |
+
"trial_name": null,
|
| 1328 |
+
"trial_params": null
|
| 1329 |
+
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4920
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:52755d98ad2bd9ec55bf7137c74905f11d5f04d9a40c982b4c6e4d07c6bb986d
|
| 3 |
size 4920
|