End of training
Browse files- README.md +5 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Dec04_10-05-11_algo-1/events.out.tfevents.1764843426.algo-1.68.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +760 -0
README.md
CHANGED
|
@@ -3,6 +3,8 @@ library_name: transformers
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: facebook/convnext-base-224-22k
|
| 5 |
tags:
|
|
|
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
@@ -16,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# Validated_cracks_raw_dataset_359_convnext_model
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [facebook/convnext-base-224-22k](https://huggingface.co/facebook/convnext-base-224-22k) on
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 1.
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: facebook/convnext-base-224-22k
|
| 5 |
tags:
|
| 6 |
+
- image-classification
|
| 7 |
+
- vision
|
| 8 |
- generated_from_trainer
|
| 9 |
metrics:
|
| 10 |
- accuracy
|
|
|
|
| 18 |
|
| 19 |
# Validated_cracks_raw_dataset_359_convnext_model
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [facebook/convnext-base-224-22k](https://huggingface.co/facebook/convnext-base-224-22k) on the Logiroad/Validated_cracks_raw_dataset_359 dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 1.0441
|
| 24 |
+
- Accuracy: 0.6021
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_accuracy": 0.602112676056338,
|
| 4 |
+
"eval_loss": 1.0440722703933716,
|
| 5 |
+
"eval_runtime": 1.9606,
|
| 6 |
+
"eval_samples_per_second": 144.85,
|
| 7 |
+
"eval_steps_per_second": 18.361,
|
| 8 |
+
"total_flos": 2.0405460235862016e+18,
|
| 9 |
+
"train_loss": 1.012899116233543,
|
| 10 |
+
"train_runtime": 686.1726,
|
| 11 |
+
"train_samples_per_second": 37.6,
|
| 12 |
+
"train_steps_per_second": 4.722
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_accuracy": 0.602112676056338,
|
| 4 |
+
"eval_loss": 1.0440722703933716,
|
| 5 |
+
"eval_runtime": 1.9606,
|
| 6 |
+
"eval_samples_per_second": 144.85,
|
| 7 |
+
"eval_steps_per_second": 18.361
|
| 8 |
+
}
|
runs/Dec04_10-05-11_algo-1/events.out.tfevents.1764843426.algo-1.68.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d8da864b7788ffc9170c5ba74e477108e87ef51cc478902377f644b6a76c5641
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"total_flos": 2.0405460235862016e+18,
|
| 4 |
+
"train_loss": 1.012899116233543,
|
| 5 |
+
"train_runtime": 686.1726,
|
| 6 |
+
"train_samples_per_second": 37.6,
|
| 7 |
+
"train_steps_per_second": 4.722
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,760 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.602112676056338,
|
| 3 |
+
"best_model_checkpoint": "./Validated_cracks_raw_dataset_359_outputs/checkpoint-1620",
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 3240,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.46296296296296297,
|
| 13 |
+
"grad_norm": 10.873927116394043,
|
| 14 |
+
"learning_rate": 2.901234567901235e-06,
|
| 15 |
+
"loss": 1.415,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.9259259259259259,
|
| 20 |
+
"grad_norm": 10.678427696228027,
|
| 21 |
+
"learning_rate": 5.9876543209876546e-06,
|
| 22 |
+
"loss": 1.3724,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.0,
|
| 27 |
+
"eval_accuracy": 0.31338028169014087,
|
| 28 |
+
"eval_loss": 1.368980884552002,
|
| 29 |
+
"eval_runtime": 1.9715,
|
| 30 |
+
"eval_samples_per_second": 144.049,
|
| 31 |
+
"eval_steps_per_second": 18.26,
|
| 32 |
+
"step": 108
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.3888888888888888,
|
| 36 |
+
"grad_norm": 15.727510452270508,
|
| 37 |
+
"learning_rate": 9.074074074074075e-06,
|
| 38 |
+
"loss": 1.3621,
|
| 39 |
+
"step": 150
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.8518518518518519,
|
| 43 |
+
"grad_norm": 18.13157844543457,
|
| 44 |
+
"learning_rate": 9.996809987196146e-06,
|
| 45 |
+
"loss": 1.3528,
|
| 46 |
+
"step": 200
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.0,
|
| 50 |
+
"eval_accuracy": 0.4014084507042254,
|
| 51 |
+
"eval_loss": 1.3014895915985107,
|
| 52 |
+
"eval_runtime": 1.9678,
|
| 53 |
+
"eval_samples_per_second": 144.321,
|
| 54 |
+
"eval_steps_per_second": 18.294,
|
| 55 |
+
"step": 216
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 2.314814814814815,
|
| 59 |
+
"grad_norm": 11.158724784851074,
|
| 60 |
+
"learning_rate": 9.981195232868493e-06,
|
| 61 |
+
"loss": 1.2858,
|
| 62 |
+
"step": 250
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.7777777777777777,
|
| 66 |
+
"grad_norm": 27.68938446044922,
|
| 67 |
+
"learning_rate": 9.952610423187516e-06,
|
| 68 |
+
"loss": 1.2839,
|
| 69 |
+
"step": 300
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 3.0,
|
| 73 |
+
"eval_accuracy": 0.43309859154929575,
|
| 74 |
+
"eval_loss": 1.2375338077545166,
|
| 75 |
+
"eval_runtime": 1.9547,
|
| 76 |
+
"eval_samples_per_second": 145.29,
|
| 77 |
+
"eval_steps_per_second": 18.417,
|
| 78 |
+
"step": 324
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 3.240740740740741,
|
| 82 |
+
"grad_norm": 16.059484481811523,
|
| 83 |
+
"learning_rate": 9.91208532978737e-06,
|
| 84 |
+
"loss": 1.2499,
|
| 85 |
+
"step": 350
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 3.7037037037037037,
|
| 89 |
+
"grad_norm": 10.013123512268066,
|
| 90 |
+
"learning_rate": 9.85807175279907e-06,
|
| 91 |
+
"loss": 1.2608,
|
| 92 |
+
"step": 400
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 4.0,
|
| 96 |
+
"eval_accuracy": 0.49295774647887325,
|
| 97 |
+
"eval_loss": 1.197479486465454,
|
| 98 |
+
"eval_runtime": 1.947,
|
| 99 |
+
"eval_samples_per_second": 145.866,
|
| 100 |
+
"eval_steps_per_second": 18.49,
|
| 101 |
+
"step": 432
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 4.166666666666667,
|
| 105 |
+
"grad_norm": 24.753084182739258,
|
| 106 |
+
"learning_rate": 9.791408709849578e-06,
|
| 107 |
+
"loss": 1.2172,
|
| 108 |
+
"step": 450
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 4.62962962962963,
|
| 112 |
+
"grad_norm": 12.23658275604248,
|
| 113 |
+
"learning_rate": 9.712269778427157e-06,
|
| 114 |
+
"loss": 1.2178,
|
| 115 |
+
"step": 500
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 5.0,
|
| 119 |
+
"eval_accuracy": 0.5246478873239436,
|
| 120 |
+
"eval_loss": 1.1367548704147339,
|
| 121 |
+
"eval_runtime": 1.9388,
|
| 122 |
+
"eval_samples_per_second": 146.485,
|
| 123 |
+
"eval_steps_per_second": 18.568,
|
| 124 |
+
"step": 540
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 5.092592592592593,
|
| 128 |
+
"grad_norm": 15.365643501281738,
|
| 129 |
+
"learning_rate": 9.620861020786583e-06,
|
| 130 |
+
"loss": 1.2052,
|
| 131 |
+
"step": 550
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 5.555555555555555,
|
| 135 |
+
"grad_norm": 12.787246704101562,
|
| 136 |
+
"learning_rate": 9.517420447403444e-06,
|
| 137 |
+
"loss": 1.1624,
|
| 138 |
+
"step": 600
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 6.0,
|
| 142 |
+
"eval_accuracy": 0.4964788732394366,
|
| 143 |
+
"eval_loss": 1.1164129972457886,
|
| 144 |
+
"eval_runtime": 1.936,
|
| 145 |
+
"eval_samples_per_second": 146.695,
|
| 146 |
+
"eval_steps_per_second": 18.595,
|
| 147 |
+
"step": 648
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 6.018518518518518,
|
| 151 |
+
"grad_norm": 13.462636947631836,
|
| 152 |
+
"learning_rate": 9.404634840109069e-06,
|
| 153 |
+
"loss": 1.166,
|
| 154 |
+
"step": 650
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 6.481481481481482,
|
| 158 |
+
"grad_norm": 15.236122131347656,
|
| 159 |
+
"learning_rate": 9.278195395448948e-06,
|
| 160 |
+
"loss": 1.1335,
|
| 161 |
+
"step": 700
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 6.944444444444445,
|
| 165 |
+
"grad_norm": 19.687719345092773,
|
| 166 |
+
"learning_rate": 9.140616369122732e-06,
|
| 167 |
+
"loss": 1.1108,
|
| 168 |
+
"step": 750
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 7.0,
|
| 172 |
+
"eval_accuracy": 0.528169014084507,
|
| 173 |
+
"eval_loss": 1.106345772743225,
|
| 174 |
+
"eval_runtime": 1.9599,
|
| 175 |
+
"eval_samples_per_second": 144.907,
|
| 176 |
+
"eval_steps_per_second": 18.368,
|
| 177 |
+
"step": 756
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 7.407407407407407,
|
| 181 |
+
"grad_norm": 16.47871971130371,
|
| 182 |
+
"learning_rate": 8.992255989929632e-06,
|
| 183 |
+
"loss": 1.0708,
|
| 184 |
+
"step": 800
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 7.87037037037037,
|
| 188 |
+
"grad_norm": 19.686279296875,
|
| 189 |
+
"learning_rate": 8.833500559197024e-06,
|
| 190 |
+
"loss": 1.1028,
|
| 191 |
+
"step": 850
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 8.0,
|
| 195 |
+
"eval_accuracy": 0.5316901408450704,
|
| 196 |
+
"eval_loss": 1.1173583269119263,
|
| 197 |
+
"eval_runtime": 1.9473,
|
| 198 |
+
"eval_samples_per_second": 145.843,
|
| 199 |
+
"eval_steps_per_second": 18.487,
|
| 200 |
+
"step": 864
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 8.333333333333334,
|
| 204 |
+
"grad_norm": 37.79273986816406,
|
| 205 |
+
"learning_rate": 8.664763444927562e-06,
|
| 206 |
+
"loss": 1.0619,
|
| 207 |
+
"step": 900
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 8.796296296296296,
|
| 211 |
+
"grad_norm": 16.420129776000977,
|
| 212 |
+
"learning_rate": 8.486484005469977e-06,
|
| 213 |
+
"loss": 1.1023,
|
| 214 |
+
"step": 950
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 9.0,
|
| 218 |
+
"eval_accuracy": 0.545774647887324,
|
| 219 |
+
"eval_loss": 1.1122897863388062,
|
| 220 |
+
"eval_runtime": 1.9381,
|
| 221 |
+
"eval_samples_per_second": 146.538,
|
| 222 |
+
"eval_steps_per_second": 18.575,
|
| 223 |
+
"step": 972
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 9.25925925925926,
|
| 227 |
+
"grad_norm": 25.430509567260742,
|
| 228 |
+
"learning_rate": 8.299126445516126e-06,
|
| 229 |
+
"loss": 1.1072,
|
| 230 |
+
"step": 1000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 9.722222222222221,
|
| 234 |
+
"grad_norm": 35.4229850769043,
|
| 235 |
+
"learning_rate": 8.103178607403005e-06,
|
| 236 |
+
"loss": 1.0572,
|
| 237 |
+
"step": 1050
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 10.0,
|
| 241 |
+
"eval_accuracy": 0.573943661971831,
|
| 242 |
+
"eval_loss": 1.0754574537277222,
|
| 243 |
+
"eval_runtime": 1.967,
|
| 244 |
+
"eval_samples_per_second": 144.38,
|
| 245 |
+
"eval_steps_per_second": 18.302,
|
| 246 |
+
"step": 1080
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 10.185185185185185,
|
| 250 |
+
"grad_norm": 14.466697692871094,
|
| 251 |
+
"learning_rate": 7.899150700867014e-06,
|
| 252 |
+
"loss": 1.0489,
|
| 253 |
+
"step": 1100
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 10.648148148148149,
|
| 257 |
+
"grad_norm": 21.01825714111328,
|
| 258 |
+
"learning_rate": 7.687573974557857e-06,
|
| 259 |
+
"loss": 0.9874,
|
| 260 |
+
"step": 1150
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 11.0,
|
| 264 |
+
"eval_accuracy": 0.5598591549295775,
|
| 265 |
+
"eval_loss": 1.095212459564209,
|
| 266 |
+
"eval_runtime": 1.9552,
|
| 267 |
+
"eval_samples_per_second": 145.255,
|
| 268 |
+
"eval_steps_per_second": 18.413,
|
| 269 |
+
"step": 1188
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 11.11111111111111,
|
| 273 |
+
"grad_norm": 16.48180389404297,
|
| 274 |
+
"learning_rate": 7.4689993327712765e-06,
|
| 275 |
+
"loss": 1.0212,
|
| 276 |
+
"step": 1200
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 11.574074074074074,
|
| 280 |
+
"grad_norm": 18.663524627685547,
|
| 281 |
+
"learning_rate": 7.243995901002312e-06,
|
| 282 |
+
"loss": 1.0132,
|
| 283 |
+
"step": 1250
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 12.0,
|
| 287 |
+
"eval_accuracy": 0.5774647887323944,
|
| 288 |
+
"eval_loss": 1.0767173767089844,
|
| 289 |
+
"eval_runtime": 1.9611,
|
| 290 |
+
"eval_samples_per_second": 144.815,
|
| 291 |
+
"eval_steps_per_second": 18.357,
|
| 292 |
+
"step": 1296
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 12.037037037037036,
|
| 296 |
+
"grad_norm": 24.026588439941406,
|
| 297 |
+
"learning_rate": 7.013149544054148e-06,
|
| 298 |
+
"loss": 1.0111,
|
| 299 |
+
"step": 1300
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 12.5,
|
| 303 |
+
"grad_norm": 20.475515365600586,
|
| 304 |
+
"learning_rate": 6.777061340561082e-06,
|
| 305 |
+
"loss": 1.005,
|
| 306 |
+
"step": 1350
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 12.962962962962964,
|
| 310 |
+
"grad_norm": 49.801326751708984,
|
| 311 |
+
"learning_rate": 6.5363460178976524e-06,
|
| 312 |
+
"loss": 0.9898,
|
| 313 |
+
"step": 1400
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 13.0,
|
| 317 |
+
"eval_accuracy": 0.5880281690140845,
|
| 318 |
+
"eval_loss": 1.061557412147522,
|
| 319 |
+
"eval_runtime": 1.9611,
|
| 320 |
+
"eval_samples_per_second": 144.818,
|
| 321 |
+
"eval_steps_per_second": 18.357,
|
| 322 |
+
"step": 1404
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 13.425925925925926,
|
| 326 |
+
"grad_norm": 38.209232330322266,
|
| 327 |
+
"learning_rate": 6.291630351549136e-06,
|
| 328 |
+
"loss": 0.9332,
|
| 329 |
+
"step": 1450
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 13.88888888888889,
|
| 333 |
+
"grad_norm": 16.34885025024414,
|
| 334 |
+
"learning_rate": 6.043551533111121e-06,
|
| 335 |
+
"loss": 1.0182,
|
| 336 |
+
"step": 1500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 14.0,
|
| 340 |
+
"eval_accuracy": 0.5809859154929577,
|
| 341 |
+
"eval_loss": 1.0419806241989136,
|
| 342 |
+
"eval_runtime": 1.9564,
|
| 343 |
+
"eval_samples_per_second": 145.165,
|
| 344 |
+
"eval_steps_per_second": 18.401,
|
| 345 |
+
"step": 1512
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 14.351851851851851,
|
| 349 |
+
"grad_norm": 21.03826904296875,
|
| 350 |
+
"learning_rate": 5.792755511167572e-06,
|
| 351 |
+
"loss": 0.9279,
|
| 352 |
+
"step": 1550
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 14.814814814814815,
|
| 356 |
+
"grad_norm": 34.984375,
|
| 357 |
+
"learning_rate": 5.544968491028696e-06,
|
| 358 |
+
"loss": 0.9889,
|
| 359 |
+
"step": 1600
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 15.0,
|
| 363 |
+
"eval_accuracy": 0.602112676056338,
|
| 364 |
+
"eval_loss": 1.0440722703933716,
|
| 365 |
+
"eval_runtime": 1.9414,
|
| 366 |
+
"eval_samples_per_second": 146.284,
|
| 367 |
+
"eval_steps_per_second": 18.543,
|
| 368 |
+
"step": 1620
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 15.277777777777779,
|
| 372 |
+
"grad_norm": 16.456146240234375,
|
| 373 |
+
"learning_rate": 5.290724144552379e-06,
|
| 374 |
+
"loss": 0.9673,
|
| 375 |
+
"step": 1650
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 15.74074074074074,
|
| 379 |
+
"grad_norm": 27.009721755981445,
|
| 380 |
+
"learning_rate": 5.035722809427533e-06,
|
| 381 |
+
"loss": 0.9446,
|
| 382 |
+
"step": 1700
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 16.0,
|
| 386 |
+
"eval_accuracy": 0.602112676056338,
|
| 387 |
+
"eval_loss": 1.0512064695358276,
|
| 388 |
+
"eval_runtime": 2.126,
|
| 389 |
+
"eval_samples_per_second": 133.583,
|
| 390 |
+
"eval_steps_per_second": 16.933,
|
| 391 |
+
"step": 1728
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 16.203703703703702,
|
| 395 |
+
"grad_norm": 12.319013595581055,
|
| 396 |
+
"learning_rate": 4.780628459113764e-06,
|
| 397 |
+
"loss": 0.9024,
|
| 398 |
+
"step": 1750
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 16.666666666666668,
|
| 402 |
+
"grad_norm": 20.682802200317383,
|
| 403 |
+
"learning_rate": 4.526105309263983e-06,
|
| 404 |
+
"loss": 0.9519,
|
| 405 |
+
"step": 1800
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 17.0,
|
| 409 |
+
"eval_accuracy": 0.5704225352112676,
|
| 410 |
+
"eval_loss": 1.0736767053604126,
|
| 411 |
+
"eval_runtime": 2.1743,
|
| 412 |
+
"eval_samples_per_second": 130.614,
|
| 413 |
+
"eval_steps_per_second": 16.557,
|
| 414 |
+
"step": 1836
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 17.12962962962963,
|
| 418 |
+
"grad_norm": 24.291671752929688,
|
| 419 |
+
"learning_rate": 4.272816088237135e-06,
|
| 420 |
+
"loss": 1.003,
|
| 421 |
+
"step": 1850
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 17.59259259259259,
|
| 425 |
+
"grad_norm": 13.384744644165039,
|
| 426 |
+
"learning_rate": 4.021420311483538e-06,
|
| 427 |
+
"loss": 0.9458,
|
| 428 |
+
"step": 1900
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 18.0,
|
| 432 |
+
"eval_accuracy": 0.5669014084507042,
|
| 433 |
+
"eval_loss": 1.0471343994140625,
|
| 434 |
+
"eval_runtime": 1.9519,
|
| 435 |
+
"eval_samples_per_second": 145.497,
|
| 436 |
+
"eval_steps_per_second": 18.443,
|
| 437 |
+
"step": 1944
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 18.055555555555557,
|
| 441 |
+
"grad_norm": 15.239733695983887,
|
| 442 |
+
"learning_rate": 3.7725725642960047e-06,
|
| 443 |
+
"loss": 0.9798,
|
| 444 |
+
"step": 1950
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 18.51851851851852,
|
| 448 |
+
"grad_norm": 24.09659767150879,
|
| 449 |
+
"learning_rate": 3.526920797398148e-06,
|
| 450 |
+
"loss": 0.9346,
|
| 451 |
+
"step": 2000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 18.98148148148148,
|
| 455 |
+
"grad_norm": 23.59765625,
|
| 456 |
+
"learning_rate": 3.2851046398077705e-06,
|
| 457 |
+
"loss": 0.9347,
|
| 458 |
+
"step": 2050
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 19.0,
|
| 462 |
+
"eval_accuracy": 0.5845070422535211,
|
| 463 |
+
"eval_loss": 1.051329255104065,
|
| 464 |
+
"eval_runtime": 1.9643,
|
| 465 |
+
"eval_samples_per_second": 144.583,
|
| 466 |
+
"eval_steps_per_second": 18.327,
|
| 467 |
+
"step": 2052
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 19.444444444444443,
|
| 471 |
+
"grad_norm": 22.252914428710938,
|
| 472 |
+
"learning_rate": 3.0477537333683815e-06,
|
| 473 |
+
"loss": 0.8767,
|
| 474 |
+
"step": 2100
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 19.90740740740741,
|
| 478 |
+
"grad_norm": 32.37730026245117,
|
| 479 |
+
"learning_rate": 2.815486093285317e-06,
|
| 480 |
+
"loss": 0.8863,
|
| 481 |
+
"step": 2150
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 20.0,
|
| 485 |
+
"eval_accuracy": 0.5950704225352113,
|
| 486 |
+
"eval_loss": 1.0428193807601929,
|
| 487 |
+
"eval_runtime": 1.9503,
|
| 488 |
+
"eval_samples_per_second": 145.616,
|
| 489 |
+
"eval_steps_per_second": 18.458,
|
| 490 |
+
"step": 2160
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 20.37037037037037,
|
| 494 |
+
"grad_norm": 18.95844841003418,
|
| 495 |
+
"learning_rate": 2.5889064989353797e-06,
|
| 496 |
+
"loss": 0.9081,
|
| 497 |
+
"step": 2200
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 20.833333333333332,
|
| 501 |
+
"grad_norm": 84.69548034667969,
|
| 502 |
+
"learning_rate": 2.3686049191399614e-06,
|
| 503 |
+
"loss": 0.8507,
|
| 504 |
+
"step": 2250
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 21.0,
|
| 508 |
+
"eval_accuracy": 0.5950704225352113,
|
| 509 |
+
"eval_loss": 1.0526705980300903,
|
| 510 |
+
"eval_runtime": 1.9472,
|
| 511 |
+
"eval_samples_per_second": 145.852,
|
| 512 |
+
"eval_steps_per_second": 18.488,
|
| 513 |
+
"step": 2268
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 21.296296296296298,
|
| 517 |
+
"grad_norm": 34.8240852355957,
|
| 518 |
+
"learning_rate": 2.155154976001948e-06,
|
| 519 |
+
"loss": 0.9308,
|
| 520 |
+
"step": 2300
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 21.75925925925926,
|
| 524 |
+
"grad_norm": 39.87207794189453,
|
| 525 |
+
"learning_rate": 1.949112451306282e-06,
|
| 526 |
+
"loss": 0.8712,
|
| 527 |
+
"step": 2350
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 22.0,
|
| 531 |
+
"eval_accuracy": 0.5915492957746479,
|
| 532 |
+
"eval_loss": 1.056043267250061,
|
| 533 |
+
"eval_runtime": 1.9559,
|
| 534 |
+
"eval_samples_per_second": 145.204,
|
| 535 |
+
"eval_steps_per_second": 18.406,
|
| 536 |
+
"step": 2376
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 22.22222222222222,
|
| 540 |
+
"grad_norm": 15.800605773925781,
|
| 541 |
+
"learning_rate": 1.7510138393732029e-06,
|
| 542 |
+
"loss": 0.8973,
|
| 543 |
+
"step": 2400
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 22.685185185185187,
|
| 547 |
+
"grad_norm": 13.362822532653809,
|
| 548 |
+
"learning_rate": 1.5613749501322705e-06,
|
| 549 |
+
"loss": 0.8857,
|
| 550 |
+
"step": 2450
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 23.0,
|
| 554 |
+
"eval_accuracy": 0.5880281690140845,
|
| 555 |
+
"eval_loss": 1.0447120666503906,
|
| 556 |
+
"eval_runtime": 1.9485,
|
| 557 |
+
"eval_samples_per_second": 145.752,
|
| 558 |
+
"eval_steps_per_second": 18.476,
|
| 559 |
+
"step": 2484
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 23.14814814814815,
|
| 563 |
+
"grad_norm": 26.928611755371094,
|
| 564 |
+
"learning_rate": 1.3806895660544805e-06,
|
| 565 |
+
"loss": 0.8446,
|
| 566 |
+
"step": 2500
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 23.61111111111111,
|
| 570 |
+
"grad_norm": 52.092063903808594,
|
| 571 |
+
"learning_rate": 1.2094281564395628e-06,
|
| 572 |
+
"loss": 0.8848,
|
| 573 |
+
"step": 2550
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 24.0,
|
| 577 |
+
"eval_accuracy": 0.5950704225352113,
|
| 578 |
+
"eval_loss": 1.051204800605774,
|
| 579 |
+
"eval_runtime": 1.9468,
|
| 580 |
+
"eval_samples_per_second": 145.879,
|
| 581 |
+
"eval_steps_per_second": 18.492,
|
| 582 |
+
"step": 2592
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 24.074074074074073,
|
| 586 |
+
"grad_norm": 22.29905128479004,
|
| 587 |
+
"learning_rate": 1.0480366524062041e-06,
|
| 588 |
+
"loss": 0.8778,
|
| 589 |
+
"step": 2600
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 24.537037037037038,
|
| 593 |
+
"grad_norm": 23.949947357177734,
|
| 594 |
+
"learning_rate": 8.969352857748842e-07,
|
| 595 |
+
"loss": 0.8672,
|
| 596 |
+
"step": 2650
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 25.0,
|
| 600 |
+
"grad_norm": 31.42759895324707,
|
| 601 |
+
"learning_rate": 7.565174948666382e-07,
|
| 602 |
+
"loss": 0.904,
|
| 603 |
+
"step": 2700
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 25.0,
|
| 607 |
+
"eval_accuracy": 0.5845070422535211,
|
| 608 |
+
"eval_loss": 1.0513179302215576,
|
| 609 |
+
"eval_runtime": 1.9542,
|
| 610 |
+
"eval_samples_per_second": 145.331,
|
| 611 |
+
"eval_steps_per_second": 18.422,
|
| 612 |
+
"step": 2700
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 25.462962962962962,
|
| 616 |
+
"grad_norm": 20.711061477661133,
|
| 617 |
+
"learning_rate": 6.271489000668418e-07,
|
| 618 |
+
"loss": 0.7932,
|
| 619 |
+
"step": 2750
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 25.925925925925927,
|
| 623 |
+
"grad_norm": 22.23466682434082,
|
| 624 |
+
"learning_rate": 5.091663518214407e-07,
|
| 625 |
+
"loss": 0.943,
|
| 626 |
+
"step": 2800
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 26.0,
|
| 630 |
+
"eval_accuracy": 0.5880281690140845,
|
| 631 |
+
"eval_loss": 1.04801344871521,
|
| 632 |
+
"eval_runtime": 1.9702,
|
| 633 |
+
"eval_samples_per_second": 144.148,
|
| 634 |
+
"eval_steps_per_second": 18.272,
|
| 635 |
+
"step": 2808
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 26.38888888888889,
|
| 639 |
+
"grad_norm": 18.16496467590332,
|
| 640 |
+
"learning_rate": 4.0287705354446147e-07,
|
| 641 |
+
"loss": 0.9108,
|
| 642 |
+
"step": 2850
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 26.85185185185185,
|
| 646 |
+
"grad_norm": 50.363128662109375,
|
| 647 |
+
"learning_rate": 3.085577617205765e-07,
|
| 648 |
+
"loss": 0.862,
|
| 649 |
+
"step": 2900
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 27.0,
|
| 653 |
+
"eval_accuracy": 0.5880281690140845,
|
| 654 |
+
"eval_loss": 1.047555685043335,
|
| 655 |
+
"eval_runtime": 1.9557,
|
| 656 |
+
"eval_samples_per_second": 145.219,
|
| 657 |
+
"eval_steps_per_second": 18.408,
|
| 658 |
+
"step": 2916
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 27.314814814814813,
|
| 662 |
+
"grad_norm": 69.7729721069336,
|
| 663 |
+
"learning_rate": 2.2645406528550407e-07,
|
| 664 |
+
"loss": 0.8412,
|
| 665 |
+
"step": 2950
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 27.77777777777778,
|
| 669 |
+
"grad_norm": 20.13194465637207,
|
| 670 |
+
"learning_rate": 1.5677974616058856e-07,
|
| 671 |
+
"loss": 0.864,
|
| 672 |
+
"step": 3000
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 28.0,
|
| 676 |
+
"eval_accuracy": 0.5880281690140845,
|
| 677 |
+
"eval_loss": 1.046897292137146,
|
| 678 |
+
"eval_runtime": 1.954,
|
| 679 |
+
"eval_samples_per_second": 145.344,
|
| 680 |
+
"eval_steps_per_second": 18.424,
|
| 681 |
+
"step": 3024
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 28.24074074074074,
|
| 685 |
+
"grad_norm": 29.41867446899414,
|
| 686 |
+
"learning_rate": 9.971622260661007e-08,
|
| 687 |
+
"loss": 0.8656,
|
| 688 |
+
"step": 3050
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 28.703703703703702,
|
| 692 |
+
"grad_norm": 19.62738609313965,
|
| 693 |
+
"learning_rate": 5.541207684621908e-08,
|
| 694 |
+
"loss": 0.8879,
|
| 695 |
+
"step": 3100
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 29.0,
|
| 699 |
+
"eval_accuracy": 0.5880281690140845,
|
| 700 |
+
"eval_loss": 1.04678475856781,
|
| 701 |
+
"eval_runtime": 1.9435,
|
| 702 |
+
"eval_samples_per_second": 146.127,
|
| 703 |
+
"eval_steps_per_second": 18.523,
|
| 704 |
+
"step": 3132
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 29.166666666666668,
|
| 708 |
+
"grad_norm": 18.350698471069336,
|
| 709 |
+
"learning_rate": 2.398266818496864e-08,
|
| 710 |
+
"loss": 0.8594,
|
| 711 |
+
"step": 3150
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 29.62962962962963,
|
| 715 |
+
"grad_norm": 28.054824829101562,
|
| 716 |
+
"learning_rate": 5.509832638314061e-09,
|
| 717 |
+
"loss": 0.8099,
|
| 718 |
+
"step": 3200
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 30.0,
|
| 722 |
+
"eval_accuracy": 0.5880281690140845,
|
| 723 |
+
"eval_loss": 1.0467545986175537,
|
| 724 |
+
"eval_runtime": 3.2936,
|
| 725 |
+
"eval_samples_per_second": 86.227,
|
| 726 |
+
"eval_steps_per_second": 10.93,
|
| 727 |
+
"step": 3240
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 30.0,
|
| 731 |
+
"step": 3240,
|
| 732 |
+
"total_flos": 2.0405460235862016e+18,
|
| 733 |
+
"train_loss": 1.012899116233543,
|
| 734 |
+
"train_runtime": 686.1726,
|
| 735 |
+
"train_samples_per_second": 37.6,
|
| 736 |
+
"train_steps_per_second": 4.722
|
| 737 |
+
}
|
| 738 |
+
],
|
| 739 |
+
"logging_steps": 50,
|
| 740 |
+
"max_steps": 3240,
|
| 741 |
+
"num_input_tokens_seen": 0,
|
| 742 |
+
"num_train_epochs": 30,
|
| 743 |
+
"save_steps": 500,
|
| 744 |
+
"stateful_callbacks": {
|
| 745 |
+
"TrainerControl": {
|
| 746 |
+
"args": {
|
| 747 |
+
"should_epoch_stop": false,
|
| 748 |
+
"should_evaluate": false,
|
| 749 |
+
"should_log": false,
|
| 750 |
+
"should_save": true,
|
| 751 |
+
"should_training_stop": true
|
| 752 |
+
},
|
| 753 |
+
"attributes": {}
|
| 754 |
+
}
|
| 755 |
+
},
|
| 756 |
+
"total_flos": 2.0405460235862016e+18,
|
| 757 |
+
"train_batch_size": 8,
|
| 758 |
+
"trial_name": null,
|
| 759 |
+
"trial_params": null
|
| 760 |
+
}
|