End of training
Browse files- README.md +24 -6
- all_results.json +16 -0
- eval_results.json +11 -0
- logs/events.out.tfevents.1675349089.serv-3334.808655.5 +3 -0
- train_results.json +8 -0
- trainer_state.json +807 -0
README.md
CHANGED
|
@@ -1,13 +1,31 @@
|
|
| 1 |
---
|
|
|
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
tags:
|
| 4 |
- generated_from_trainer
|
|
|
|
|
|
|
| 5 |
metrics:
|
| 6 |
- accuracy
|
| 7 |
- f1
|
| 8 |
model-index:
|
| 9 |
- name: mobilebert_sa_GLUE_Experiment_logit_kd_data_aug_mrpc
|
| 10 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
---
|
| 12 |
|
| 13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -15,12 +33,12 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# mobilebert_sa_GLUE_Experiment_logit_kd_data_aug_mrpc
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [google/mobilebert-uncased](https://huggingface.co/google/mobilebert-uncased) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
-
- F1: 0.
|
| 23 |
-
- Combined Score: 0.
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 1 |
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
license: apache-2.0
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
| 7 |
+
datasets:
|
| 8 |
+
- glue
|
| 9 |
metrics:
|
| 10 |
- accuracy
|
| 11 |
- f1
|
| 12 |
model-index:
|
| 13 |
- name: mobilebert_sa_GLUE_Experiment_logit_kd_data_aug_mrpc
|
| 14 |
+
results:
|
| 15 |
+
- task:
|
| 16 |
+
name: Text Classification
|
| 17 |
+
type: text-classification
|
| 18 |
+
dataset:
|
| 19 |
+
name: GLUE MRPC
|
| 20 |
+
type: glue
|
| 21 |
+
args: mrpc
|
| 22 |
+
metrics:
|
| 23 |
+
- name: Accuracy
|
| 24 |
+
type: accuracy
|
| 25 |
+
value: 0.9877450980392157
|
| 26 |
+
- name: F1
|
| 27 |
+
type: f1
|
| 28 |
+
value: 0.9911190053285969
|
| 29 |
---
|
| 30 |
|
| 31 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 33 |
|
| 34 |
# mobilebert_sa_GLUE_Experiment_logit_kd_data_aug_mrpc
|
| 35 |
|
| 36 |
+
This model is a fine-tuned version of [google/mobilebert-uncased](https://huggingface.co/google/mobilebert-uncased) on the GLUE MRPC dataset.
|
| 37 |
It achieves the following results on the evaluation set:
|
| 38 |
+
- Loss: 0.1256
|
| 39 |
+
- Accuracy: 0.9877
|
| 40 |
+
- F1: 0.9911
|
| 41 |
+
- Combined Score: 0.9894
|
| 42 |
|
| 43 |
## Model description
|
| 44 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 46.0,
|
| 3 |
+
"eval_accuracy": 0.9877450980392157,
|
| 4 |
+
"eval_combined_score": 0.9894320516839064,
|
| 5 |
+
"eval_f1": 0.9911190053285969,
|
| 6 |
+
"eval_loss": 0.1256455034017563,
|
| 7 |
+
"eval_runtime": 0.7942,
|
| 8 |
+
"eval_samples": 408,
|
| 9 |
+
"eval_samples_per_second": 513.756,
|
| 10 |
+
"eval_steps_per_second": 5.037,
|
| 11 |
+
"train_loss": 0.19228331322046327,
|
| 12 |
+
"train_runtime": 51035.534,
|
| 13 |
+
"train_samples": 250736,
|
| 14 |
+
"train_samples_per_second": 245.648,
|
| 15 |
+
"train_steps_per_second": 1.919
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 46.0,
|
| 3 |
+
"eval_accuracy": 0.9877450980392157,
|
| 4 |
+
"eval_combined_score": 0.9894320516839064,
|
| 5 |
+
"eval_f1": 0.9911190053285969,
|
| 6 |
+
"eval_loss": 0.1256455034017563,
|
| 7 |
+
"eval_runtime": 0.7942,
|
| 8 |
+
"eval_samples": 408,
|
| 9 |
+
"eval_samples_per_second": 513.756,
|
| 10 |
+
"eval_steps_per_second": 5.037
|
| 11 |
+
}
|
logs/events.out.tfevents.1675349089.serv-3334.808655.5
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1bbb5174e6b011d58f9fe88b25651ec80cd4be2f5205cfbe71b4340f2f319ff7
|
| 3 |
+
size 475
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 46.0,
|
| 3 |
+
"train_loss": 0.19228331322046327,
|
| 4 |
+
"train_runtime": 51035.534,
|
| 5 |
+
"train_samples": 250736,
|
| 6 |
+
"train_samples_per_second": 245.648,
|
| 7 |
+
"train_steps_per_second": 1.919
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,807 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.1256455034017563,
|
| 3 |
+
"best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_logit_kd_data_aug_mrpc/checkpoint-80319",
|
| 4 |
+
"epoch": 46.0,
|
| 5 |
+
"global_step": 90114,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.0,
|
| 12 |
+
"learning_rate": 4.9e-05,
|
| 13 |
+
"loss": 0.2964,
|
| 14 |
+
"step": 1959
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.0,
|
| 18 |
+
"eval_accuracy": 0.9607843137254902,
|
| 19 |
+
"eval_combined_score": 0.9663076498204916,
|
| 20 |
+
"eval_f1": 0.9718309859154929,
|
| 21 |
+
"eval_loss": 0.20259901881217957,
|
| 22 |
+
"eval_runtime": 0.7823,
|
| 23 |
+
"eval_samples_per_second": 521.547,
|
| 24 |
+
"eval_steps_per_second": 5.113,
|
| 25 |
+
"step": 1959
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 2.0,
|
| 29 |
+
"learning_rate": 4.8e-05,
|
| 30 |
+
"loss": 0.2307,
|
| 31 |
+
"step": 3918
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 2.0,
|
| 35 |
+
"eval_accuracy": 0.9705882352941176,
|
| 36 |
+
"eval_combined_score": 0.9747678018575852,
|
| 37 |
+
"eval_f1": 0.9789473684210526,
|
| 38 |
+
"eval_loss": 0.19426828622817993,
|
| 39 |
+
"eval_runtime": 0.7765,
|
| 40 |
+
"eval_samples_per_second": 525.442,
|
| 41 |
+
"eval_steps_per_second": 5.151,
|
| 42 |
+
"step": 3918
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 3.0,
|
| 46 |
+
"learning_rate": 4.7e-05,
|
| 47 |
+
"loss": 0.2221,
|
| 48 |
+
"step": 5877
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 3.0,
|
| 52 |
+
"eval_accuracy": 0.9803921568627451,
|
| 53 |
+
"eval_combined_score": 0.9831038798498122,
|
| 54 |
+
"eval_f1": 0.9858156028368794,
|
| 55 |
+
"eval_loss": 0.18736393749713898,
|
| 56 |
+
"eval_runtime": 0.773,
|
| 57 |
+
"eval_samples_per_second": 527.78,
|
| 58 |
+
"eval_steps_per_second": 5.174,
|
| 59 |
+
"step": 5877
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 4.0,
|
| 63 |
+
"learning_rate": 4.600000000000001e-05,
|
| 64 |
+
"loss": 0.2163,
|
| 65 |
+
"step": 7836
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 4.0,
|
| 69 |
+
"eval_accuracy": 0.9852941176470589,
|
| 70 |
+
"eval_combined_score": 0.9873279098873593,
|
| 71 |
+
"eval_f1": 0.9893617021276596,
|
| 72 |
+
"eval_loss": 0.17034168541431427,
|
| 73 |
+
"eval_runtime": 0.7757,
|
| 74 |
+
"eval_samples_per_second": 526.002,
|
| 75 |
+
"eval_steps_per_second": 5.157,
|
| 76 |
+
"step": 7836
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 5.0,
|
| 80 |
+
"learning_rate": 4.5e-05,
|
| 81 |
+
"loss": 0.2115,
|
| 82 |
+
"step": 9795
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 5.0,
|
| 86 |
+
"eval_accuracy": 0.9852941176470589,
|
| 87 |
+
"eval_combined_score": 0.9873279098873593,
|
| 88 |
+
"eval_f1": 0.9893617021276596,
|
| 89 |
+
"eval_loss": 0.18048794567584991,
|
| 90 |
+
"eval_runtime": 0.7764,
|
| 91 |
+
"eval_samples_per_second": 525.485,
|
| 92 |
+
"eval_steps_per_second": 5.152,
|
| 93 |
+
"step": 9795
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 6.0,
|
| 97 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 98 |
+
"loss": 0.2071,
|
| 99 |
+
"step": 11754
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 6.0,
|
| 103 |
+
"eval_accuracy": 0.9803921568627451,
|
| 104 |
+
"eval_combined_score": 0.9831289406221853,
|
| 105 |
+
"eval_f1": 0.9858657243816255,
|
| 106 |
+
"eval_loss": 0.16820354759693146,
|
| 107 |
+
"eval_runtime": 0.7809,
|
| 108 |
+
"eval_samples_per_second": 522.481,
|
| 109 |
+
"eval_steps_per_second": 5.122,
|
| 110 |
+
"step": 11754
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 7.0,
|
| 114 |
+
"learning_rate": 4.3e-05,
|
| 115 |
+
"loss": 0.2036,
|
| 116 |
+
"step": 13713
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 7.0,
|
| 120 |
+
"eval_accuracy": 0.9877450980392157,
|
| 121 |
+
"eval_combined_score": 0.9894320516839064,
|
| 122 |
+
"eval_f1": 0.9911190053285969,
|
| 123 |
+
"eval_loss": 0.15831029415130615,
|
| 124 |
+
"eval_runtime": 0.78,
|
| 125 |
+
"eval_samples_per_second": 523.105,
|
| 126 |
+
"eval_steps_per_second": 5.128,
|
| 127 |
+
"step": 13713
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 8.0,
|
| 131 |
+
"learning_rate": 4.2e-05,
|
| 132 |
+
"loss": 0.2007,
|
| 133 |
+
"step": 15672
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 8.0,
|
| 137 |
+
"eval_accuracy": 0.9926470588235294,
|
| 138 |
+
"eval_combined_score": 0.9936497326203209,
|
| 139 |
+
"eval_f1": 0.9946524064171123,
|
| 140 |
+
"eval_loss": 0.16275052726268768,
|
| 141 |
+
"eval_runtime": 0.7788,
|
| 142 |
+
"eval_samples_per_second": 523.893,
|
| 143 |
+
"eval_steps_per_second": 5.136,
|
| 144 |
+
"step": 15672
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 9.0,
|
| 148 |
+
"learning_rate": 4.1e-05,
|
| 149 |
+
"loss": 0.1985,
|
| 150 |
+
"step": 17631
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 9.0,
|
| 154 |
+
"eval_accuracy": 0.9852941176470589,
|
| 155 |
+
"eval_combined_score": 0.9873279098873593,
|
| 156 |
+
"eval_f1": 0.9893617021276596,
|
| 157 |
+
"eval_loss": 0.15478737652301788,
|
| 158 |
+
"eval_runtime": 0.7768,
|
| 159 |
+
"eval_samples_per_second": 525.26,
|
| 160 |
+
"eval_steps_per_second": 5.15,
|
| 161 |
+
"step": 17631
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 10.0,
|
| 165 |
+
"learning_rate": 4e-05,
|
| 166 |
+
"loss": 0.1965,
|
| 167 |
+
"step": 19590
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 10.0,
|
| 171 |
+
"eval_accuracy": 0.9877450980392157,
|
| 172 |
+
"eval_combined_score": 0.9894162210338681,
|
| 173 |
+
"eval_f1": 0.9910873440285204,
|
| 174 |
+
"eval_loss": 0.15827800333499908,
|
| 175 |
+
"eval_runtime": 0.7788,
|
| 176 |
+
"eval_samples_per_second": 523.897,
|
| 177 |
+
"eval_steps_per_second": 5.136,
|
| 178 |
+
"step": 19590
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 11.0,
|
| 182 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 183 |
+
"loss": 0.195,
|
| 184 |
+
"step": 21549
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 11.0,
|
| 188 |
+
"eval_accuracy": 0.9901960784313726,
|
| 189 |
+
"eval_combined_score": 0.9915138098250054,
|
| 190 |
+
"eval_f1": 0.992831541218638,
|
| 191 |
+
"eval_loss": 0.1527128666639328,
|
| 192 |
+
"eval_runtime": 0.7814,
|
| 193 |
+
"eval_samples_per_second": 522.15,
|
| 194 |
+
"eval_steps_per_second": 5.119,
|
| 195 |
+
"step": 21549
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 12.0,
|
| 199 |
+
"learning_rate": 3.8e-05,
|
| 200 |
+
"loss": 0.1938,
|
| 201 |
+
"step": 23508
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 12.0,
|
| 205 |
+
"eval_accuracy": 0.9901960784313726,
|
| 206 |
+
"eval_combined_score": 0.9915266106442577,
|
| 207 |
+
"eval_f1": 0.9928571428571428,
|
| 208 |
+
"eval_loss": 0.15122443437576294,
|
| 209 |
+
"eval_runtime": 0.7775,
|
| 210 |
+
"eval_samples_per_second": 524.756,
|
| 211 |
+
"eval_steps_per_second": 5.145,
|
| 212 |
+
"step": 23508
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 13.0,
|
| 216 |
+
"learning_rate": 3.7e-05,
|
| 217 |
+
"loss": 0.1926,
|
| 218 |
+
"step": 25467
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 13.0,
|
| 222 |
+
"eval_accuracy": 0.9950980392156863,
|
| 223 |
+
"eval_combined_score": 0.9957633053221289,
|
| 224 |
+
"eval_f1": 0.9964285714285714,
|
| 225 |
+
"eval_loss": 0.1425897628068924,
|
| 226 |
+
"eval_runtime": 0.7755,
|
| 227 |
+
"eval_samples_per_second": 526.144,
|
| 228 |
+
"eval_steps_per_second": 5.158,
|
| 229 |
+
"step": 25467
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 14.0,
|
| 233 |
+
"learning_rate": 3.6e-05,
|
| 234 |
+
"loss": 0.1917,
|
| 235 |
+
"step": 27426
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 14.0,
|
| 239 |
+
"eval_accuracy": 0.9950980392156863,
|
| 240 |
+
"eval_combined_score": 0.9957633053221289,
|
| 241 |
+
"eval_f1": 0.9964285714285714,
|
| 242 |
+
"eval_loss": 0.14360927045345306,
|
| 243 |
+
"eval_runtime": 0.7791,
|
| 244 |
+
"eval_samples_per_second": 523.666,
|
| 245 |
+
"eval_steps_per_second": 5.134,
|
| 246 |
+
"step": 27426
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 15.0,
|
| 250 |
+
"learning_rate": 3.5e-05,
|
| 251 |
+
"loss": 0.191,
|
| 252 |
+
"step": 29385
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 15.0,
|
| 256 |
+
"eval_accuracy": 0.9926470588235294,
|
| 257 |
+
"eval_combined_score": 0.9936401662632853,
|
| 258 |
+
"eval_f1": 0.9946332737030412,
|
| 259 |
+
"eval_loss": 0.1502537876367569,
|
| 260 |
+
"eval_runtime": 0.7823,
|
| 261 |
+
"eval_samples_per_second": 521.529,
|
| 262 |
+
"eval_steps_per_second": 5.113,
|
| 263 |
+
"step": 29385
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 16.0,
|
| 267 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 268 |
+
"loss": 0.1901,
|
| 269 |
+
"step": 31344
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 16.0,
|
| 273 |
+
"eval_accuracy": 0.9950980392156863,
|
| 274 |
+
"eval_combined_score": 0.9957633053221289,
|
| 275 |
+
"eval_f1": 0.9964285714285714,
|
| 276 |
+
"eval_loss": 0.14606404304504395,
|
| 277 |
+
"eval_runtime": 0.7757,
|
| 278 |
+
"eval_samples_per_second": 525.986,
|
| 279 |
+
"eval_steps_per_second": 5.157,
|
| 280 |
+
"step": 31344
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 17.0,
|
| 284 |
+
"learning_rate": 3.3e-05,
|
| 285 |
+
"loss": 0.1894,
|
| 286 |
+
"step": 33303
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 17.0,
|
| 290 |
+
"eval_accuracy": 0.9975490196078431,
|
| 291 |
+
"eval_combined_score": 0.9978800554210951,
|
| 292 |
+
"eval_f1": 0.998211091234347,
|
| 293 |
+
"eval_loss": 0.1498469114303589,
|
| 294 |
+
"eval_runtime": 0.7758,
|
| 295 |
+
"eval_samples_per_second": 525.884,
|
| 296 |
+
"eval_steps_per_second": 5.156,
|
| 297 |
+
"step": 33303
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 18.0,
|
| 301 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 302 |
+
"loss": 0.1888,
|
| 303 |
+
"step": 35262
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 18.0,
|
| 307 |
+
"eval_accuracy": 0.9901960784313726,
|
| 308 |
+
"eval_combined_score": 0.9915266106442577,
|
| 309 |
+
"eval_f1": 0.9928571428571428,
|
| 310 |
+
"eval_loss": 0.14023421704769135,
|
| 311 |
+
"eval_runtime": 0.7766,
|
| 312 |
+
"eval_samples_per_second": 525.375,
|
| 313 |
+
"eval_steps_per_second": 5.151,
|
| 314 |
+
"step": 35262
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 19.0,
|
| 318 |
+
"learning_rate": 3.1e-05,
|
| 319 |
+
"loss": 0.1882,
|
| 320 |
+
"step": 37221
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 19.0,
|
| 324 |
+
"eval_accuracy": 0.9926470588235294,
|
| 325 |
+
"eval_combined_score": 0.9936401662632853,
|
| 326 |
+
"eval_f1": 0.9946332737030412,
|
| 327 |
+
"eval_loss": 0.14202910661697388,
|
| 328 |
+
"eval_runtime": 0.7794,
|
| 329 |
+
"eval_samples_per_second": 523.449,
|
| 330 |
+
"eval_steps_per_second": 5.132,
|
| 331 |
+
"step": 37221
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 20.0,
|
| 335 |
+
"learning_rate": 3e-05,
|
| 336 |
+
"loss": 0.1876,
|
| 337 |
+
"step": 39180
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 20.0,
|
| 341 |
+
"eval_accuracy": 0.9901960784313726,
|
| 342 |
+
"eval_combined_score": 0.9915393203544763,
|
| 343 |
+
"eval_f1": 0.9928825622775801,
|
| 344 |
+
"eval_loss": 0.13458134233951569,
|
| 345 |
+
"eval_runtime": 0.7781,
|
| 346 |
+
"eval_samples_per_second": 524.385,
|
| 347 |
+
"eval_steps_per_second": 5.141,
|
| 348 |
+
"step": 39180
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 21.0,
|
| 352 |
+
"learning_rate": 2.9e-05,
|
| 353 |
+
"loss": 0.1871,
|
| 354 |
+
"step": 41139
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 21.0,
|
| 358 |
+
"eval_accuracy": 0.9950980392156863,
|
| 359 |
+
"eval_combined_score": 0.9957633053221289,
|
| 360 |
+
"eval_f1": 0.9964285714285714,
|
| 361 |
+
"eval_loss": 0.13956362009048462,
|
| 362 |
+
"eval_runtime": 0.7768,
|
| 363 |
+
"eval_samples_per_second": 525.249,
|
| 364 |
+
"eval_steps_per_second": 5.15,
|
| 365 |
+
"step": 41139
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 22.0,
|
| 369 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 370 |
+
"loss": 0.1867,
|
| 371 |
+
"step": 43098
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 22.0,
|
| 375 |
+
"eval_accuracy": 0.9950980392156863,
|
| 376 |
+
"eval_combined_score": 0.9957633053221289,
|
| 377 |
+
"eval_f1": 0.9964285714285714,
|
| 378 |
+
"eval_loss": 0.14428994059562683,
|
| 379 |
+
"eval_runtime": 0.7783,
|
| 380 |
+
"eval_samples_per_second": 524.192,
|
| 381 |
+
"eval_steps_per_second": 5.139,
|
| 382 |
+
"step": 43098
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 23.0,
|
| 386 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 387 |
+
"loss": 0.1862,
|
| 388 |
+
"step": 45057
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 23.0,
|
| 392 |
+
"eval_accuracy": 0.9926470588235294,
|
| 393 |
+
"eval_combined_score": 0.9936497326203209,
|
| 394 |
+
"eval_f1": 0.9946524064171123,
|
| 395 |
+
"eval_loss": 0.1346110701560974,
|
| 396 |
+
"eval_runtime": 0.7827,
|
| 397 |
+
"eval_samples_per_second": 521.253,
|
| 398 |
+
"eval_steps_per_second": 5.11,
|
| 399 |
+
"step": 45057
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 24.0,
|
| 403 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 404 |
+
"loss": 0.1857,
|
| 405 |
+
"step": 47016
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 24.0,
|
| 409 |
+
"eval_accuracy": 0.9950980392156863,
|
| 410 |
+
"eval_combined_score": 0.9957633053221289,
|
| 411 |
+
"eval_f1": 0.9964285714285714,
|
| 412 |
+
"eval_loss": 0.13611076772212982,
|
| 413 |
+
"eval_runtime": 0.7925,
|
| 414 |
+
"eval_samples_per_second": 514.847,
|
| 415 |
+
"eval_steps_per_second": 5.048,
|
| 416 |
+
"step": 47016
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 25.0,
|
| 420 |
+
"learning_rate": 2.5e-05,
|
| 421 |
+
"loss": 0.1854,
|
| 422 |
+
"step": 48975
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 25.0,
|
| 426 |
+
"eval_accuracy": 0.9926470588235294,
|
| 427 |
+
"eval_combined_score": 0.9936497326203209,
|
| 428 |
+
"eval_f1": 0.9946524064171123,
|
| 429 |
+
"eval_loss": 0.13179393112659454,
|
| 430 |
+
"eval_runtime": 0.7822,
|
| 431 |
+
"eval_samples_per_second": 521.594,
|
| 432 |
+
"eval_steps_per_second": 5.114,
|
| 433 |
+
"step": 48975
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 26.0,
|
| 437 |
+
"learning_rate": 2.4e-05,
|
| 438 |
+
"loss": 0.185,
|
| 439 |
+
"step": 50934
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 26.0,
|
| 443 |
+
"eval_accuracy": 0.9901960784313726,
|
| 444 |
+
"eval_combined_score": 0.9915393203544763,
|
| 445 |
+
"eval_f1": 0.9928825622775801,
|
| 446 |
+
"eval_loss": 0.13099727034568787,
|
| 447 |
+
"eval_runtime": 0.851,
|
| 448 |
+
"eval_samples_per_second": 479.464,
|
| 449 |
+
"eval_steps_per_second": 4.701,
|
| 450 |
+
"step": 50934
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 27.0,
|
| 454 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 455 |
+
"loss": 0.1846,
|
| 456 |
+
"step": 52893
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 27.0,
|
| 460 |
+
"eval_accuracy": 0.9926470588235294,
|
| 461 |
+
"eval_combined_score": 0.9936497326203209,
|
| 462 |
+
"eval_f1": 0.9946524064171123,
|
| 463 |
+
"eval_loss": 0.13022534549236298,
|
| 464 |
+
"eval_runtime": 0.7818,
|
| 465 |
+
"eval_samples_per_second": 521.85,
|
| 466 |
+
"eval_steps_per_second": 5.116,
|
| 467 |
+
"step": 52893
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 28.0,
|
| 471 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 472 |
+
"loss": 0.1842,
|
| 473 |
+
"step": 54852
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 28.0,
|
| 477 |
+
"eval_accuracy": 0.9950980392156863,
|
| 478 |
+
"eval_combined_score": 0.9957633053221289,
|
| 479 |
+
"eval_f1": 0.9964285714285714,
|
| 480 |
+
"eval_loss": 0.13285598158836365,
|
| 481 |
+
"eval_runtime": 0.7818,
|
| 482 |
+
"eval_samples_per_second": 521.866,
|
| 483 |
+
"eval_steps_per_second": 5.116,
|
| 484 |
+
"step": 54852
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 29.0,
|
| 488 |
+
"learning_rate": 2.1e-05,
|
| 489 |
+
"loss": 0.1839,
|
| 490 |
+
"step": 56811
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 29.0,
|
| 494 |
+
"eval_accuracy": 0.9901960784313726,
|
| 495 |
+
"eval_combined_score": 0.9915393203544763,
|
| 496 |
+
"eval_f1": 0.9928825622775801,
|
| 497 |
+
"eval_loss": 0.13004697859287262,
|
| 498 |
+
"eval_runtime": 0.7761,
|
| 499 |
+
"eval_samples_per_second": 525.714,
|
| 500 |
+
"eval_steps_per_second": 5.154,
|
| 501 |
+
"step": 56811
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 30.0,
|
| 505 |
+
"learning_rate": 2e-05,
|
| 506 |
+
"loss": 0.1836,
|
| 507 |
+
"step": 58770
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 30.0,
|
| 511 |
+
"eval_accuracy": 0.9901960784313726,
|
| 512 |
+
"eval_combined_score": 0.9915393203544763,
|
| 513 |
+
"eval_f1": 0.9928825622775801,
|
| 514 |
+
"eval_loss": 0.1328008770942688,
|
| 515 |
+
"eval_runtime": 0.7782,
|
| 516 |
+
"eval_samples_per_second": 524.253,
|
| 517 |
+
"eval_steps_per_second": 5.14,
|
| 518 |
+
"step": 58770
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 31.0,
|
| 522 |
+
"learning_rate": 1.9e-05,
|
| 523 |
+
"loss": 0.1832,
|
| 524 |
+
"step": 60729
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 31.0,
|
| 528 |
+
"eval_accuracy": 0.9901960784313726,
|
| 529 |
+
"eval_combined_score": 0.9915393203544763,
|
| 530 |
+
"eval_f1": 0.9928825622775801,
|
| 531 |
+
"eval_loss": 0.1326626092195511,
|
| 532 |
+
"eval_runtime": 0.7751,
|
| 533 |
+
"eval_samples_per_second": 526.407,
|
| 534 |
+
"eval_steps_per_second": 5.161,
|
| 535 |
+
"step": 60729
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 32.0,
|
| 539 |
+
"learning_rate": 1.8e-05,
|
| 540 |
+
"loss": 0.1829,
|
| 541 |
+
"step": 62688
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 32.0,
|
| 545 |
+
"eval_accuracy": 0.9901960784313726,
|
| 546 |
+
"eval_combined_score": 0.9915393203544763,
|
| 547 |
+
"eval_f1": 0.9928825622775801,
|
| 548 |
+
"eval_loss": 0.13079187273979187,
|
| 549 |
+
"eval_runtime": 0.7779,
|
| 550 |
+
"eval_samples_per_second": 524.516,
|
| 551 |
+
"eval_steps_per_second": 5.142,
|
| 552 |
+
"step": 62688
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 33.0,
|
| 556 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 557 |
+
"loss": 0.1826,
|
| 558 |
+
"step": 64647
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 33.0,
|
| 562 |
+
"eval_accuracy": 0.9901960784313726,
|
| 563 |
+
"eval_combined_score": 0.9915393203544763,
|
| 564 |
+
"eval_f1": 0.9928825622775801,
|
| 565 |
+
"eval_loss": 0.12873521447181702,
|
| 566 |
+
"eval_runtime": 0.7782,
|
| 567 |
+
"eval_samples_per_second": 524.295,
|
| 568 |
+
"eval_steps_per_second": 5.14,
|
| 569 |
+
"step": 64647
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 34.0,
|
| 573 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 574 |
+
"loss": 0.1824,
|
| 575 |
+
"step": 66606
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 34.0,
|
| 579 |
+
"eval_accuracy": 0.9926470588235294,
|
| 580 |
+
"eval_combined_score": 0.9936497326203209,
|
| 581 |
+
"eval_f1": 0.9946524064171123,
|
| 582 |
+
"eval_loss": 0.1308642029762268,
|
| 583 |
+
"eval_runtime": 0.7782,
|
| 584 |
+
"eval_samples_per_second": 524.283,
|
| 585 |
+
"eval_steps_per_second": 5.14,
|
| 586 |
+
"step": 66606
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 35.0,
|
| 590 |
+
"learning_rate": 1.5e-05,
|
| 591 |
+
"loss": 0.1821,
|
| 592 |
+
"step": 68565
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 35.0,
|
| 596 |
+
"eval_accuracy": 0.9926470588235294,
|
| 597 |
+
"eval_combined_score": 0.9936497326203209,
|
| 598 |
+
"eval_f1": 0.9946524064171123,
|
| 599 |
+
"eval_loss": 0.13088451325893402,
|
| 600 |
+
"eval_runtime": 0.7758,
|
| 601 |
+
"eval_samples_per_second": 525.926,
|
| 602 |
+
"eval_steps_per_second": 5.156,
|
| 603 |
+
"step": 68565
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 36.0,
|
| 607 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 608 |
+
"loss": 0.1818,
|
| 609 |
+
"step": 70524
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 36.0,
|
| 613 |
+
"eval_accuracy": 0.9901960784313726,
|
| 614 |
+
"eval_combined_score": 0.9915393203544763,
|
| 615 |
+
"eval_f1": 0.9928825622775801,
|
| 616 |
+
"eval_loss": 0.1271485686302185,
|
| 617 |
+
"eval_runtime": 0.7809,
|
| 618 |
+
"eval_samples_per_second": 522.489,
|
| 619 |
+
"eval_steps_per_second": 5.122,
|
| 620 |
+
"step": 70524
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 37.0,
|
| 624 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 625 |
+
"loss": 0.1816,
|
| 626 |
+
"step": 72483
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 37.0,
|
| 630 |
+
"eval_accuracy": 0.9877450980392157,
|
| 631 |
+
"eval_combined_score": 0.9894320516839064,
|
| 632 |
+
"eval_f1": 0.9911190053285969,
|
| 633 |
+
"eval_loss": 0.127828910946846,
|
| 634 |
+
"eval_runtime": 0.7755,
|
| 635 |
+
"eval_samples_per_second": 526.078,
|
| 636 |
+
"eval_steps_per_second": 5.158,
|
| 637 |
+
"step": 72483
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 38.0,
|
| 641 |
+
"learning_rate": 1.2e-05,
|
| 642 |
+
"loss": 0.1813,
|
| 643 |
+
"step": 74442
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 38.0,
|
| 647 |
+
"eval_accuracy": 0.9901960784313726,
|
| 648 |
+
"eval_combined_score": 0.9915393203544763,
|
| 649 |
+
"eval_f1": 0.9928825622775801,
|
| 650 |
+
"eval_loss": 0.1280445158481598,
|
| 651 |
+
"eval_runtime": 0.7777,
|
| 652 |
+
"eval_samples_per_second": 524.647,
|
| 653 |
+
"eval_steps_per_second": 5.144,
|
| 654 |
+
"step": 74442
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 39.0,
|
| 658 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 659 |
+
"loss": 0.1811,
|
| 660 |
+
"step": 76401
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 39.0,
|
| 664 |
+
"eval_accuracy": 0.9901960784313726,
|
| 665 |
+
"eval_combined_score": 0.9915393203544763,
|
| 666 |
+
"eval_f1": 0.9928825622775801,
|
| 667 |
+
"eval_loss": 0.12891501188278198,
|
| 668 |
+
"eval_runtime": 0.7798,
|
| 669 |
+
"eval_samples_per_second": 523.193,
|
| 670 |
+
"eval_steps_per_second": 5.129,
|
| 671 |
+
"step": 76401
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 40.0,
|
| 675 |
+
"learning_rate": 1e-05,
|
| 676 |
+
"loss": 0.1809,
|
| 677 |
+
"step": 78360
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 40.0,
|
| 681 |
+
"eval_accuracy": 0.9877450980392157,
|
| 682 |
+
"eval_combined_score": 0.9894320516839064,
|
| 683 |
+
"eval_f1": 0.9911190053285969,
|
| 684 |
+
"eval_loss": 0.12895023822784424,
|
| 685 |
+
"eval_runtime": 0.7761,
|
| 686 |
+
"eval_samples_per_second": 525.732,
|
| 687 |
+
"eval_steps_per_second": 5.154,
|
| 688 |
+
"step": 78360
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 41.0,
|
| 692 |
+
"learning_rate": 9e-06,
|
| 693 |
+
"loss": 0.1807,
|
| 694 |
+
"step": 80319
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 41.0,
|
| 698 |
+
"eval_accuracy": 0.9877450980392157,
|
| 699 |
+
"eval_combined_score": 0.9894320516839064,
|
| 700 |
+
"eval_f1": 0.9911190053285969,
|
| 701 |
+
"eval_loss": 0.1256455034017563,
|
| 702 |
+
"eval_runtime": 0.7742,
|
| 703 |
+
"eval_samples_per_second": 526.998,
|
| 704 |
+
"eval_steps_per_second": 5.167,
|
| 705 |
+
"step": 80319
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 42.0,
|
| 709 |
+
"learning_rate": 8.000000000000001e-06,
|
| 710 |
+
"loss": 0.1805,
|
| 711 |
+
"step": 82278
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 42.0,
|
| 715 |
+
"eval_accuracy": 0.9926470588235294,
|
| 716 |
+
"eval_combined_score": 0.9936497326203209,
|
| 717 |
+
"eval_f1": 0.9946524064171123,
|
| 718 |
+
"eval_loss": 0.12676292657852173,
|
| 719 |
+
"eval_runtime": 0.7808,
|
| 720 |
+
"eval_samples_per_second": 522.525,
|
| 721 |
+
"eval_steps_per_second": 5.123,
|
| 722 |
+
"step": 82278
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 43.0,
|
| 726 |
+
"learning_rate": 7.000000000000001e-06,
|
| 727 |
+
"loss": 0.1803,
|
| 728 |
+
"step": 84237
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 43.0,
|
| 732 |
+
"eval_accuracy": 0.9926470588235294,
|
| 733 |
+
"eval_combined_score": 0.9936497326203209,
|
| 734 |
+
"eval_f1": 0.9946524064171123,
|
| 735 |
+
"eval_loss": 0.12742425501346588,
|
| 736 |
+
"eval_runtime": 0.7744,
|
| 737 |
+
"eval_samples_per_second": 526.85,
|
| 738 |
+
"eval_steps_per_second": 5.165,
|
| 739 |
+
"step": 84237
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 44.0,
|
| 743 |
+
"learning_rate": 6e-06,
|
| 744 |
+
"loss": 0.1801,
|
| 745 |
+
"step": 86196
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 44.0,
|
| 749 |
+
"eval_accuracy": 0.9926470588235294,
|
| 750 |
+
"eval_combined_score": 0.9936497326203209,
|
| 751 |
+
"eval_f1": 0.9946524064171123,
|
| 752 |
+
"eval_loss": 0.12773118913173676,
|
| 753 |
+
"eval_runtime": 0.7768,
|
| 754 |
+
"eval_samples_per_second": 525.233,
|
| 755 |
+
"eval_steps_per_second": 5.149,
|
| 756 |
+
"step": 86196
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 45.0,
|
| 760 |
+
"learning_rate": 5e-06,
|
| 761 |
+
"loss": 0.1799,
|
| 762 |
+
"step": 88155
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 45.0,
|
| 766 |
+
"eval_accuracy": 0.9926470588235294,
|
| 767 |
+
"eval_combined_score": 0.9936497326203209,
|
| 768 |
+
"eval_f1": 0.9946524064171123,
|
| 769 |
+
"eval_loss": 0.12637551128864288,
|
| 770 |
+
"eval_runtime": 0.7756,
|
| 771 |
+
"eval_samples_per_second": 526.076,
|
| 772 |
+
"eval_steps_per_second": 5.158,
|
| 773 |
+
"step": 88155
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 46.0,
|
| 777 |
+
"learning_rate": 4.000000000000001e-06,
|
| 778 |
+
"loss": 0.1797,
|
| 779 |
+
"step": 90114
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 46.0,
|
| 783 |
+
"eval_accuracy": 0.9901960784313726,
|
| 784 |
+
"eval_combined_score": 0.9915393203544763,
|
| 785 |
+
"eval_f1": 0.9928825622775801,
|
| 786 |
+
"eval_loss": 0.12735696136951447,
|
| 787 |
+
"eval_runtime": 0.7805,
|
| 788 |
+
"eval_samples_per_second": 522.735,
|
| 789 |
+
"eval_steps_per_second": 5.125,
|
| 790 |
+
"step": 90114
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 46.0,
|
| 794 |
+
"step": 90114,
|
| 795 |
+
"total_flos": 3.6163552369495245e+17,
|
| 796 |
+
"train_loss": 0.19228331322046327,
|
| 797 |
+
"train_runtime": 51035.534,
|
| 798 |
+
"train_samples_per_second": 245.648,
|
| 799 |
+
"train_steps_per_second": 1.919
|
| 800 |
+
}
|
| 801 |
+
],
|
| 802 |
+
"max_steps": 97950,
|
| 803 |
+
"num_train_epochs": 50,
|
| 804 |
+
"total_flos": 3.6163552369495245e+17,
|
| 805 |
+
"trial_name": null,
|
| 806 |
+
"trial_params": null
|
| 807 |
+
}
|