🍻 cheers
Browse files- README.md +4 -3
- all_results.json +11 -11
- eval_results.json +6 -6
- runs/Oct01_10-59-17_a59caee1d103/events.out.tfevents.1727781816.a59caee1d103.4704.8 +3 -0
- train_results.json +6 -6
- trainer_state.json +653 -45
README.md
CHANGED
|
@@ -3,6 +3,7 @@ library_name: transformers
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: google/vit-base-patch16-224-in21k
|
| 5 |
tags:
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
@@ -16,10 +17,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# finetuned-fake-food
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: google/vit-base-patch16-224-in21k
|
| 5 |
tags:
|
| 6 |
+
- image-classification
|
| 7 |
- generated_from_trainer
|
| 8 |
metrics:
|
| 9 |
- accuracy
|
|
|
|
| 17 |
|
| 18 |
# finetuned-fake-food
|
| 19 |
|
| 20 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the indian_food_images dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.6574
|
| 23 |
+
- Accuracy: 0.6164
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second":
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.6164383561643836,
|
| 4 |
+
"eval_loss": 0.6574238538742065,
|
| 5 |
+
"eval_runtime": 6.2004,
|
| 6 |
+
"eval_samples_per_second": 23.547,
|
| 7 |
+
"eval_steps_per_second": 3.064,
|
| 8 |
+
"total_flos": 6.400838342165299e+17,
|
| 9 |
+
"train_loss": 0.6831480086282725,
|
| 10 |
+
"train_runtime": 1157.4221,
|
| 11 |
+
"train_samples_per_second": 7.137,
|
| 12 |
+
"train_steps_per_second": 3.568
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.6164383561643836,
|
| 4 |
+
"eval_loss": 0.6574238538742065,
|
| 5 |
+
"eval_runtime": 6.2004,
|
| 6 |
+
"eval_samples_per_second": 23.547,
|
| 7 |
+
"eval_steps_per_second": 3.064
|
| 8 |
}
|
runs/Oct01_10-59-17_a59caee1d103/events.out.tfevents.1727781816.a59caee1d103.4704.8
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a80a22aad7ade61395653bb941d35ce87de78d580c4adbd6d2feb81c7353e4e
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 6.400838342165299e+17,
|
| 4 |
+
"train_loss": 0.6831480086282725,
|
| 5 |
+
"train_runtime": 1157.4221,
|
| 6 |
+
"train_samples_per_second": 7.137,
|
| 7 |
+
"train_steps_per_second": 3.568
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,75 +1,683 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "finetuned-fake-food/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 100,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch":
|
| 13 |
-
"grad_norm": 0.
|
| 14 |
-
"learning_rate": 0.
|
| 15 |
-
"loss": 0.
|
| 16 |
"step": 100
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch":
|
| 20 |
-
"eval_accuracy": 0.
|
| 21 |
-
"eval_loss": 0.
|
| 22 |
-
"eval_runtime":
|
| 23 |
-
"eval_samples_per_second":
|
| 24 |
-
"eval_steps_per_second":
|
| 25 |
"step": 100
|
| 26 |
},
|
| 27 |
{
|
| 28 |
-
"epoch":
|
| 29 |
-
"grad_norm":
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss": 0.
|
| 32 |
"step": 200
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch":
|
| 36 |
-
"eval_accuracy": 0.
|
| 37 |
-
"eval_loss": 0.
|
| 38 |
-
"eval_runtime":
|
| 39 |
-
"eval_samples_per_second":
|
| 40 |
-
"eval_steps_per_second":
|
| 41 |
"step": 200
|
| 42 |
},
|
| 43 |
{
|
| 44 |
-
"epoch":
|
| 45 |
-
"grad_norm":
|
| 46 |
-
"learning_rate": 0.
|
| 47 |
-
"loss": 0.
|
| 48 |
"step": 300
|
| 49 |
},
|
| 50 |
{
|
| 51 |
-
"epoch":
|
| 52 |
-
"eval_accuracy": 0.
|
| 53 |
-
"eval_loss": 0.
|
| 54 |
-
"eval_runtime":
|
| 55 |
-
"eval_samples_per_second":
|
| 56 |
-
"eval_steps_per_second":
|
| 57 |
"step": 300
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch":
|
| 61 |
-
"
|
| 62 |
-
"
|
| 63 |
-
"
|
| 64 |
-
"
|
| 65 |
-
|
| 66 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 67 |
}
|
| 68 |
],
|
| 69 |
"logging_steps": 100,
|
| 70 |
-
"max_steps":
|
| 71 |
"num_input_tokens_seen": 0,
|
| 72 |
-
"num_train_epochs":
|
| 73 |
"save_steps": 100,
|
| 74 |
"stateful_callbacks": {
|
| 75 |
"TrainerControl": {
|
|
@@ -83,8 +691,8 @@
|
|
| 83 |
"attributes": {}
|
| 84 |
}
|
| 85 |
},
|
| 86 |
-
"total_flos":
|
| 87 |
-
"train_batch_size":
|
| 88 |
"trial_name": null,
|
| 89 |
"trial_params": null
|
| 90 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.6574238538742065,
|
| 3 |
+
"best_model_checkpoint": "finetuned-fake-food/checkpoint-4000",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 4130,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.24213075060532688,
|
| 13 |
+
"grad_norm": 0.15737777948379517,
|
| 14 |
+
"learning_rate": 0.00019515738498789345,
|
| 15 |
+
"loss": 0.6977,
|
| 16 |
"step": 100
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.24213075060532688,
|
| 20 |
+
"eval_accuracy": 0.5821917808219178,
|
| 21 |
+
"eval_loss": 0.6804767847061157,
|
| 22 |
+
"eval_runtime": 5.525,
|
| 23 |
+
"eval_samples_per_second": 26.425,
|
| 24 |
+
"eval_steps_per_second": 3.439,
|
| 25 |
"step": 100
|
| 26 |
},
|
| 27 |
{
|
| 28 |
+
"epoch": 0.48426150121065376,
|
| 29 |
+
"grad_norm": 0.03558634594082832,
|
| 30 |
+
"learning_rate": 0.00019031476997578695,
|
| 31 |
+
"loss": 0.6956,
|
| 32 |
"step": 200
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 0.48426150121065376,
|
| 36 |
+
"eval_accuracy": 0.4178082191780822,
|
| 37 |
+
"eval_loss": 0.6935968399047852,
|
| 38 |
+
"eval_runtime": 5.2465,
|
| 39 |
+
"eval_samples_per_second": 27.828,
|
| 40 |
+
"eval_steps_per_second": 3.621,
|
| 41 |
"step": 200
|
| 42 |
},
|
| 43 |
{
|
| 44 |
+
"epoch": 0.7263922518159807,
|
| 45 |
+
"grad_norm": 1.5009288787841797,
|
| 46 |
+
"learning_rate": 0.0001854721549636804,
|
| 47 |
+
"loss": 0.6795,
|
| 48 |
"step": 300
|
| 49 |
},
|
| 50 |
{
|
| 51 |
+
"epoch": 0.7263922518159807,
|
| 52 |
+
"eval_accuracy": 0.6506849315068494,
|
| 53 |
+
"eval_loss": 0.6734184622764587,
|
| 54 |
+
"eval_runtime": 5.9372,
|
| 55 |
+
"eval_samples_per_second": 24.591,
|
| 56 |
+
"eval_steps_per_second": 3.2,
|
| 57 |
"step": 300
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 0.9685230024213075,
|
| 61 |
+
"grad_norm": 0.23967677354812622,
|
| 62 |
+
"learning_rate": 0.00018062953995157384,
|
| 63 |
+
"loss": 0.7061,
|
| 64 |
+
"step": 400
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.9685230024213075,
|
| 68 |
+
"eval_accuracy": 0.5821917808219178,
|
| 69 |
+
"eval_loss": 0.6760488152503967,
|
| 70 |
+
"eval_runtime": 5.7531,
|
| 71 |
+
"eval_samples_per_second": 25.378,
|
| 72 |
+
"eval_steps_per_second": 3.303,
|
| 73 |
+
"step": 400
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.2106537530266344,
|
| 77 |
+
"grad_norm": 2.081388235092163,
|
| 78 |
+
"learning_rate": 0.00017578692493946732,
|
| 79 |
+
"loss": 0.6941,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.2106537530266344,
|
| 84 |
+
"eval_accuracy": 0.5821917808219178,
|
| 85 |
+
"eval_loss": 0.6746240854263306,
|
| 86 |
+
"eval_runtime": 5.812,
|
| 87 |
+
"eval_samples_per_second": 25.121,
|
| 88 |
+
"eval_steps_per_second": 3.269,
|
| 89 |
+
"step": 500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.4527845036319613,
|
| 93 |
+
"grad_norm": 0.4285804331302643,
|
| 94 |
+
"learning_rate": 0.0001709443099273608,
|
| 95 |
+
"loss": 0.6898,
|
| 96 |
+
"step": 600
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.4527845036319613,
|
| 100 |
+
"eval_accuracy": 0.6027397260273972,
|
| 101 |
+
"eval_loss": 0.6674954891204834,
|
| 102 |
+
"eval_runtime": 6.1547,
|
| 103 |
+
"eval_samples_per_second": 23.722,
|
| 104 |
+
"eval_steps_per_second": 3.087,
|
| 105 |
+
"step": 600
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.694915254237288,
|
| 109 |
+
"grad_norm": 0.12052281200885773,
|
| 110 |
+
"learning_rate": 0.00016610169491525423,
|
| 111 |
+
"loss": 0.6956,
|
| 112 |
+
"step": 700
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.694915254237288,
|
| 116 |
+
"eval_accuracy": 0.5753424657534246,
|
| 117 |
+
"eval_loss": 0.684603750705719,
|
| 118 |
+
"eval_runtime": 6.0144,
|
| 119 |
+
"eval_samples_per_second": 24.275,
|
| 120 |
+
"eval_steps_per_second": 3.159,
|
| 121 |
+
"step": 700
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.937046004842615,
|
| 125 |
+
"grad_norm": 0.3585425913333893,
|
| 126 |
+
"learning_rate": 0.0001612590799031477,
|
| 127 |
+
"loss": 0.6847,
|
| 128 |
+
"step": 800
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.937046004842615,
|
| 132 |
+
"eval_accuracy": 0.5821917808219178,
|
| 133 |
+
"eval_loss": 0.6745873093605042,
|
| 134 |
+
"eval_runtime": 5.8427,
|
| 135 |
+
"eval_samples_per_second": 24.988,
|
| 136 |
+
"eval_steps_per_second": 3.252,
|
| 137 |
+
"step": 800
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.179176755447942,
|
| 141 |
+
"grad_norm": 1.2445541620254517,
|
| 142 |
+
"learning_rate": 0.00015641646489104115,
|
| 143 |
+
"loss": 0.6949,
|
| 144 |
+
"step": 900
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.179176755447942,
|
| 148 |
+
"eval_accuracy": 0.589041095890411,
|
| 149 |
+
"eval_loss": 0.6779718399047852,
|
| 150 |
+
"eval_runtime": 4.8307,
|
| 151 |
+
"eval_samples_per_second": 30.223,
|
| 152 |
+
"eval_steps_per_second": 3.933,
|
| 153 |
+
"step": 900
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.4213075060532687,
|
| 157 |
+
"grad_norm": 1.429865837097168,
|
| 158 |
+
"learning_rate": 0.00015157384987893465,
|
| 159 |
+
"loss": 0.703,
|
| 160 |
+
"step": 1000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.4213075060532687,
|
| 164 |
+
"eval_accuracy": 0.5753424657534246,
|
| 165 |
+
"eval_loss": 0.6894732117652893,
|
| 166 |
+
"eval_runtime": 5.0834,
|
| 167 |
+
"eval_samples_per_second": 28.721,
|
| 168 |
+
"eval_steps_per_second": 3.738,
|
| 169 |
+
"step": 1000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 2.663438256658596,
|
| 173 |
+
"grad_norm": 1.2485073804855347,
|
| 174 |
+
"learning_rate": 0.0001467312348668281,
|
| 175 |
+
"loss": 0.6851,
|
| 176 |
+
"step": 1100
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 2.663438256658596,
|
| 180 |
+
"eval_accuracy": 0.5821917808219178,
|
| 181 |
+
"eval_loss": 0.6741558909416199,
|
| 182 |
+
"eval_runtime": 6.0858,
|
| 183 |
+
"eval_samples_per_second": 23.99,
|
| 184 |
+
"eval_steps_per_second": 3.122,
|
| 185 |
+
"step": 1100
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 2.9055690072639226,
|
| 189 |
+
"grad_norm": 0.1997382789850235,
|
| 190 |
+
"learning_rate": 0.00014188861985472154,
|
| 191 |
+
"loss": 0.6878,
|
| 192 |
+
"step": 1200
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 2.9055690072639226,
|
| 196 |
+
"eval_accuracy": 0.6301369863013698,
|
| 197 |
+
"eval_loss": 0.674239456653595,
|
| 198 |
+
"eval_runtime": 6.1499,
|
| 199 |
+
"eval_samples_per_second": 23.74,
|
| 200 |
+
"eval_steps_per_second": 3.089,
|
| 201 |
+
"step": 1200
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 3.1476997578692494,
|
| 205 |
+
"grad_norm": 0.4324168860912323,
|
| 206 |
+
"learning_rate": 0.00013704600484261504,
|
| 207 |
+
"loss": 0.68,
|
| 208 |
+
"step": 1300
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 3.1476997578692494,
|
| 212 |
+
"eval_accuracy": 0.5821917808219178,
|
| 213 |
+
"eval_loss": 0.6712663173675537,
|
| 214 |
+
"eval_runtime": 6.0266,
|
| 215 |
+
"eval_samples_per_second": 24.226,
|
| 216 |
+
"eval_steps_per_second": 3.153,
|
| 217 |
+
"step": 1300
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 3.389830508474576,
|
| 221 |
+
"grad_norm": 0.9948041439056396,
|
| 222 |
+
"learning_rate": 0.00013220338983050849,
|
| 223 |
+
"loss": 0.6728,
|
| 224 |
+
"step": 1400
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 3.389830508474576,
|
| 228 |
+
"eval_accuracy": 0.5958904109589042,
|
| 229 |
+
"eval_loss": 0.6838211417198181,
|
| 230 |
+
"eval_runtime": 6.1631,
|
| 231 |
+
"eval_samples_per_second": 23.689,
|
| 232 |
+
"eval_steps_per_second": 3.083,
|
| 233 |
+
"step": 1400
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 3.6319612590799033,
|
| 237 |
+
"grad_norm": 1.2490299940109253,
|
| 238 |
+
"learning_rate": 0.00012736077481840193,
|
| 239 |
+
"loss": 0.698,
|
| 240 |
+
"step": 1500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.6319612590799033,
|
| 244 |
+
"eval_accuracy": 0.5821917808219178,
|
| 245 |
+
"eval_loss": 0.6774668097496033,
|
| 246 |
+
"eval_runtime": 6.0857,
|
| 247 |
+
"eval_samples_per_second": 23.991,
|
| 248 |
+
"eval_steps_per_second": 3.122,
|
| 249 |
+
"step": 1500
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 3.87409200968523,
|
| 253 |
+
"grad_norm": 0.2908919155597687,
|
| 254 |
+
"learning_rate": 0.0001225181598062954,
|
| 255 |
+
"loss": 0.7033,
|
| 256 |
+
"step": 1600
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 3.87409200968523,
|
| 260 |
+
"eval_accuracy": 0.5821917808219178,
|
| 261 |
+
"eval_loss": 0.6734635829925537,
|
| 262 |
+
"eval_runtime": 5.783,
|
| 263 |
+
"eval_samples_per_second": 25.247,
|
| 264 |
+
"eval_steps_per_second": 3.286,
|
| 265 |
+
"step": 1600
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 4.116222760290557,
|
| 269 |
+
"grad_norm": 0.21098549664020538,
|
| 270 |
+
"learning_rate": 0.00011767554479418887,
|
| 271 |
+
"loss": 0.6973,
|
| 272 |
+
"step": 1700
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 4.116222760290557,
|
| 276 |
+
"eval_accuracy": 0.6232876712328768,
|
| 277 |
+
"eval_loss": 0.6803831458091736,
|
| 278 |
+
"eval_runtime": 4.6265,
|
| 279 |
+
"eval_samples_per_second": 31.557,
|
| 280 |
+
"eval_steps_per_second": 4.107,
|
| 281 |
+
"step": 1700
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 4.358353510895884,
|
| 285 |
+
"grad_norm": 0.03869936615228653,
|
| 286 |
+
"learning_rate": 0.00011283292978208233,
|
| 287 |
+
"loss": 0.6822,
|
| 288 |
+
"step": 1800
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 4.358353510895884,
|
| 292 |
+
"eval_accuracy": 0.6027397260273972,
|
| 293 |
+
"eval_loss": 0.6847726702690125,
|
| 294 |
+
"eval_runtime": 4.6717,
|
| 295 |
+
"eval_samples_per_second": 31.252,
|
| 296 |
+
"eval_steps_per_second": 4.067,
|
| 297 |
+
"step": 1800
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.600484261501211,
|
| 301 |
+
"grad_norm": 0.13196176290512085,
|
| 302 |
+
"learning_rate": 0.00010799031476997579,
|
| 303 |
+
"loss": 0.6896,
|
| 304 |
+
"step": 1900
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 4.600484261501211,
|
| 308 |
+
"eval_accuracy": 0.541095890410959,
|
| 309 |
+
"eval_loss": 0.6835151314735413,
|
| 310 |
+
"eval_runtime": 5.191,
|
| 311 |
+
"eval_samples_per_second": 28.126,
|
| 312 |
+
"eval_steps_per_second": 3.66,
|
| 313 |
+
"step": 1900
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 4.842615012106537,
|
| 317 |
+
"grad_norm": 0.5055987238883972,
|
| 318 |
+
"learning_rate": 0.00010314769975786926,
|
| 319 |
+
"loss": 0.6772,
|
| 320 |
+
"step": 2000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 4.842615012106537,
|
| 324 |
+
"eval_accuracy": 0.6095890410958904,
|
| 325 |
+
"eval_loss": 0.6753013134002686,
|
| 326 |
+
"eval_runtime": 6.1231,
|
| 327 |
+
"eval_samples_per_second": 23.844,
|
| 328 |
+
"eval_steps_per_second": 3.103,
|
| 329 |
+
"step": 2000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 5.084745762711864,
|
| 333 |
+
"grad_norm": 0.4209335148334503,
|
| 334 |
+
"learning_rate": 9.835351089588378e-05,
|
| 335 |
+
"loss": 0.6843,
|
| 336 |
+
"step": 2100
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 5.084745762711864,
|
| 340 |
+
"eval_accuracy": 0.589041095890411,
|
| 341 |
+
"eval_loss": 0.6667279601097107,
|
| 342 |
+
"eval_runtime": 4.6413,
|
| 343 |
+
"eval_samples_per_second": 31.457,
|
| 344 |
+
"eval_steps_per_second": 4.094,
|
| 345 |
+
"step": 2100
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 5.326876513317191,
|
| 349 |
+
"grad_norm": 1.4965670108795166,
|
| 350 |
+
"learning_rate": 9.351089588377724e-05,
|
| 351 |
+
"loss": 0.6898,
|
| 352 |
+
"step": 2200
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 5.326876513317191,
|
| 356 |
+
"eval_accuracy": 0.5821917808219178,
|
| 357 |
+
"eval_loss": 0.6725605726242065,
|
| 358 |
+
"eval_runtime": 6.034,
|
| 359 |
+
"eval_samples_per_second": 24.196,
|
| 360 |
+
"eval_steps_per_second": 3.149,
|
| 361 |
+
"step": 2200
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 5.5690072639225185,
|
| 365 |
+
"grad_norm": 0.4163062572479248,
|
| 366 |
+
"learning_rate": 8.86682808716707e-05,
|
| 367 |
+
"loss": 0.6868,
|
| 368 |
+
"step": 2300
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 5.5690072639225185,
|
| 372 |
+
"eval_accuracy": 0.5616438356164384,
|
| 373 |
+
"eval_loss": 0.6784049272537231,
|
| 374 |
+
"eval_runtime": 6.0533,
|
| 375 |
+
"eval_samples_per_second": 24.119,
|
| 376 |
+
"eval_steps_per_second": 3.139,
|
| 377 |
+
"step": 2300
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 5.811138014527845,
|
| 381 |
+
"grad_norm": 1.2287280559539795,
|
| 382 |
+
"learning_rate": 8.382566585956417e-05,
|
| 383 |
+
"loss": 0.6636,
|
| 384 |
+
"step": 2400
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 5.811138014527845,
|
| 388 |
+
"eval_accuracy": 0.6301369863013698,
|
| 389 |
+
"eval_loss": 0.6639688611030579,
|
| 390 |
+
"eval_runtime": 5.3104,
|
| 391 |
+
"eval_samples_per_second": 27.493,
|
| 392 |
+
"eval_steps_per_second": 3.578,
|
| 393 |
+
"step": 2400
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 6.053268765133172,
|
| 397 |
+
"grad_norm": 0.8932170867919922,
|
| 398 |
+
"learning_rate": 7.898305084745763e-05,
|
| 399 |
+
"loss": 0.6833,
|
| 400 |
+
"step": 2500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 6.053268765133172,
|
| 404 |
+
"eval_accuracy": 0.5136986301369864,
|
| 405 |
+
"eval_loss": 0.676824688911438,
|
| 406 |
+
"eval_runtime": 4.6631,
|
| 407 |
+
"eval_samples_per_second": 31.309,
|
| 408 |
+
"eval_steps_per_second": 4.074,
|
| 409 |
+
"step": 2500
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 6.295399515738499,
|
| 413 |
+
"grad_norm": 1.1837154626846313,
|
| 414 |
+
"learning_rate": 7.414043583535109e-05,
|
| 415 |
+
"loss": 0.678,
|
| 416 |
+
"step": 2600
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 6.295399515738499,
|
| 420 |
+
"eval_accuracy": 0.6232876712328768,
|
| 421 |
+
"eval_loss": 0.6652230024337769,
|
| 422 |
+
"eval_runtime": 4.933,
|
| 423 |
+
"eval_samples_per_second": 29.597,
|
| 424 |
+
"eval_steps_per_second": 3.852,
|
| 425 |
+
"step": 2600
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 6.5375302663438255,
|
| 429 |
+
"grad_norm": 1.4030615091323853,
|
| 430 |
+
"learning_rate": 6.929782082324455e-05,
|
| 431 |
+
"loss": 0.6672,
|
| 432 |
+
"step": 2700
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 6.5375302663438255,
|
| 436 |
+
"eval_accuracy": 0.547945205479452,
|
| 437 |
+
"eval_loss": 0.6735221147537231,
|
| 438 |
+
"eval_runtime": 6.0106,
|
| 439 |
+
"eval_samples_per_second": 24.29,
|
| 440 |
+
"eval_steps_per_second": 3.161,
|
| 441 |
+
"step": 2700
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 6.779661016949152,
|
| 445 |
+
"grad_norm": 0.6782599687576294,
|
| 446 |
+
"learning_rate": 6.445520581113802e-05,
|
| 447 |
+
"loss": 0.6975,
|
| 448 |
+
"step": 2800
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 6.779661016949152,
|
| 452 |
+
"eval_accuracy": 0.589041095890411,
|
| 453 |
+
"eval_loss": 0.6686810851097107,
|
| 454 |
+
"eval_runtime": 4.7582,
|
| 455 |
+
"eval_samples_per_second": 30.684,
|
| 456 |
+
"eval_steps_per_second": 3.993,
|
| 457 |
+
"step": 2800
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 7.021791767554479,
|
| 461 |
+
"grad_norm": 0.4288092255592346,
|
| 462 |
+
"learning_rate": 5.961259079903147e-05,
|
| 463 |
+
"loss": 0.6858,
|
| 464 |
+
"step": 2900
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 7.021791767554479,
|
| 468 |
+
"eval_accuracy": 0.6027397260273972,
|
| 469 |
+
"eval_loss": 0.6672346591949463,
|
| 470 |
+
"eval_runtime": 4.7612,
|
| 471 |
+
"eval_samples_per_second": 30.665,
|
| 472 |
+
"eval_steps_per_second": 3.991,
|
| 473 |
+
"step": 2900
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 7.263922518159807,
|
| 477 |
+
"grad_norm": 1.373633861541748,
|
| 478 |
+
"learning_rate": 5.4769975786924946e-05,
|
| 479 |
+
"loss": 0.6687,
|
| 480 |
+
"step": 3000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 7.263922518159807,
|
| 484 |
+
"eval_accuracy": 0.5753424657534246,
|
| 485 |
+
"eval_loss": 0.6647915840148926,
|
| 486 |
+
"eval_runtime": 4.681,
|
| 487 |
+
"eval_samples_per_second": 31.19,
|
| 488 |
+
"eval_steps_per_second": 4.059,
|
| 489 |
+
"step": 3000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 7.506053268765133,
|
| 493 |
+
"grad_norm": 0.4883480668067932,
|
| 494 |
+
"learning_rate": 4.9927360774818404e-05,
|
| 495 |
+
"loss": 0.6636,
|
| 496 |
+
"step": 3100
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 7.506053268765133,
|
| 500 |
+
"eval_accuracy": 0.5684931506849316,
|
| 501 |
+
"eval_loss": 0.6673935055732727,
|
| 502 |
+
"eval_runtime": 4.6832,
|
| 503 |
+
"eval_samples_per_second": 31.175,
|
| 504 |
+
"eval_steps_per_second": 4.057,
|
| 505 |
+
"step": 3100
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 7.74818401937046,
|
| 509 |
+
"grad_norm": 0.2553524672985077,
|
| 510 |
+
"learning_rate": 4.508474576271187e-05,
|
| 511 |
+
"loss": 0.6904,
|
| 512 |
+
"step": 3200
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 7.74818401937046,
|
| 516 |
+
"eval_accuracy": 0.5342465753424658,
|
| 517 |
+
"eval_loss": 0.6751775741577148,
|
| 518 |
+
"eval_runtime": 4.7128,
|
| 519 |
+
"eval_samples_per_second": 30.979,
|
| 520 |
+
"eval_steps_per_second": 4.032,
|
| 521 |
+
"step": 3200
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 7.990314769975787,
|
| 525 |
+
"grad_norm": 0.5203524827957153,
|
| 526 |
+
"learning_rate": 4.024213075060533e-05,
|
| 527 |
+
"loss": 0.6585,
|
| 528 |
+
"step": 3300
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 7.990314769975787,
|
| 532 |
+
"eval_accuracy": 0.5958904109589042,
|
| 533 |
+
"eval_loss": 0.7023173570632935,
|
| 534 |
+
"eval_runtime": 5.9291,
|
| 535 |
+
"eval_samples_per_second": 24.624,
|
| 536 |
+
"eval_steps_per_second": 3.205,
|
| 537 |
+
"step": 3300
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 8.232445520581114,
|
| 541 |
+
"grad_norm": 1.3221914768218994,
|
| 542 |
+
"learning_rate": 3.539951573849879e-05,
|
| 543 |
+
"loss": 0.6874,
|
| 544 |
+
"step": 3400
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 8.232445520581114,
|
| 548 |
+
"eval_accuracy": 0.5753424657534246,
|
| 549 |
+
"eval_loss": 0.6615224480628967,
|
| 550 |
+
"eval_runtime": 5.9877,
|
| 551 |
+
"eval_samples_per_second": 24.383,
|
| 552 |
+
"eval_steps_per_second": 3.173,
|
| 553 |
+
"step": 3400
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 8.474576271186441,
|
| 557 |
+
"grad_norm": 0.6332941651344299,
|
| 558 |
+
"learning_rate": 3.055690072639225e-05,
|
| 559 |
+
"loss": 0.6444,
|
| 560 |
+
"step": 3500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 8.474576271186441,
|
| 564 |
+
"eval_accuracy": 0.5205479452054794,
|
| 565 |
+
"eval_loss": 0.772119402885437,
|
| 566 |
+
"eval_runtime": 5.8976,
|
| 567 |
+
"eval_samples_per_second": 24.756,
|
| 568 |
+
"eval_steps_per_second": 3.222,
|
| 569 |
+
"step": 3500
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 8.716707021791768,
|
| 573 |
+
"grad_norm": 1.8709771633148193,
|
| 574 |
+
"learning_rate": 2.5714285714285714e-05,
|
| 575 |
+
"loss": 0.6803,
|
| 576 |
+
"step": 3600
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 8.716707021791768,
|
| 580 |
+
"eval_accuracy": 0.5821917808219178,
|
| 581 |
+
"eval_loss": 0.6808822751045227,
|
| 582 |
+
"eval_runtime": 4.743,
|
| 583 |
+
"eval_samples_per_second": 30.782,
|
| 584 |
+
"eval_steps_per_second": 4.006,
|
| 585 |
+
"step": 3600
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 8.958837772397095,
|
| 589 |
+
"grad_norm": 1.1251460313796997,
|
| 590 |
+
"learning_rate": 2.0871670702179177e-05,
|
| 591 |
+
"loss": 0.6782,
|
| 592 |
+
"step": 3700
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 8.958837772397095,
|
| 596 |
+
"eval_accuracy": 0.5821917808219178,
|
| 597 |
+
"eval_loss": 0.6638409495353699,
|
| 598 |
+
"eval_runtime": 5.3184,
|
| 599 |
+
"eval_samples_per_second": 27.452,
|
| 600 |
+
"eval_steps_per_second": 3.573,
|
| 601 |
+
"step": 3700
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 9.200968523002421,
|
| 605 |
+
"grad_norm": 0.5918538570404053,
|
| 606 |
+
"learning_rate": 1.602905569007264e-05,
|
| 607 |
+
"loss": 0.6536,
|
| 608 |
+
"step": 3800
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 9.200968523002421,
|
| 612 |
+
"eval_accuracy": 0.6232876712328768,
|
| 613 |
+
"eval_loss": 0.6606671214103699,
|
| 614 |
+
"eval_runtime": 4.6821,
|
| 615 |
+
"eval_samples_per_second": 31.183,
|
| 616 |
+
"eval_steps_per_second": 4.058,
|
| 617 |
+
"step": 3800
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 9.443099273607748,
|
| 621 |
+
"grad_norm": 1.1931524276733398,
|
| 622 |
+
"learning_rate": 1.1186440677966102e-05,
|
| 623 |
+
"loss": 0.6188,
|
| 624 |
+
"step": 3900
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 9.443099273607748,
|
| 628 |
+
"eval_accuracy": 0.5684931506849316,
|
| 629 |
+
"eval_loss": 0.7090215682983398,
|
| 630 |
+
"eval_runtime": 4.6915,
|
| 631 |
+
"eval_samples_per_second": 31.12,
|
| 632 |
+
"eval_steps_per_second": 4.05,
|
| 633 |
+
"step": 3900
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 9.685230024213075,
|
| 637 |
+
"grad_norm": 2.177264928817749,
|
| 638 |
+
"learning_rate": 6.3438256658595635e-06,
|
| 639 |
+
"loss": 0.7026,
|
| 640 |
+
"step": 4000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 9.685230024213075,
|
| 644 |
+
"eval_accuracy": 0.6164383561643836,
|
| 645 |
+
"eval_loss": 0.6574238538742065,
|
| 646 |
+
"eval_runtime": 4.7502,
|
| 647 |
+
"eval_samples_per_second": 30.736,
|
| 648 |
+
"eval_steps_per_second": 4.0,
|
| 649 |
+
"step": 4000
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 9.927360774818402,
|
| 653 |
+
"grad_norm": 0.5541784167289734,
|
| 654 |
+
"learning_rate": 1.549636803874092e-06,
|
| 655 |
+
"loss": 0.7008,
|
| 656 |
+
"step": 4100
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 9.927360774818402,
|
| 660 |
+
"eval_accuracy": 0.6095890410958904,
|
| 661 |
+
"eval_loss": 0.6576805710792542,
|
| 662 |
+
"eval_runtime": 4.7337,
|
| 663 |
+
"eval_samples_per_second": 30.843,
|
| 664 |
+
"eval_steps_per_second": 4.014,
|
| 665 |
+
"step": 4100
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 10.0,
|
| 669 |
+
"step": 4130,
|
| 670 |
+
"total_flos": 6.400838342165299e+17,
|
| 671 |
+
"train_loss": 0.6831480086282725,
|
| 672 |
+
"train_runtime": 1157.4221,
|
| 673 |
+
"train_samples_per_second": 7.137,
|
| 674 |
+
"train_steps_per_second": 3.568
|
| 675 |
}
|
| 676 |
],
|
| 677 |
"logging_steps": 100,
|
| 678 |
+
"max_steps": 4130,
|
| 679 |
"num_input_tokens_seen": 0,
|
| 680 |
+
"num_train_epochs": 10,
|
| 681 |
"save_steps": 100,
|
| 682 |
"stateful_callbacks": {
|
| 683 |
"TrainerControl": {
|
|
|
|
| 691 |
"attributes": {}
|
| 692 |
}
|
| 693 |
},
|
| 694 |
+
"total_flos": 6.400838342165299e+17,
|
| 695 |
+
"train_batch_size": 2,
|
| 696 |
"trial_name": null,
|
| 697 |
"trial_params": null
|
| 698 |
}
|