End of training
Browse files- README.md +14 -2
- all_results.json +16 -0
- eval_results.json +10 -0
- tokenizer.json +1 -6
- train_results.json +9 -0
- trainer_state.json +1729 -0
README.md
CHANGED
|
@@ -3,11 +3,23 @@ license: other
|
|
| 3 |
base_model: Qwen/Qwen1.5-4B
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
|
|
|
|
|
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
| 8 |
model-index:
|
| 9 |
- name: lmind_hotpot_train8000_eval7405_v1_recite_qa_Qwen_Qwen1.5-4B_3e-5_lora2
|
| 10 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
library_name: peft
|
| 12 |
---
|
| 13 |
|
|
@@ -16,7 +28,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# lmind_hotpot_train8000_eval7405_v1_recite_qa_Qwen_Qwen1.5-4B_3e-5_lora2
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [Qwen/Qwen1.5-4B](https://huggingface.co/Qwen/Qwen1.5-4B) on
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.9440
|
| 22 |
- Accuracy: 0.7263
|
|
|
|
| 3 |
base_model: Qwen/Qwen1.5-4B
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- tyzhu/lmind_hotpot_train8000_eval7405_v1_recite_qa
|
| 8 |
metrics:
|
| 9 |
- accuracy
|
| 10 |
model-index:
|
| 11 |
- name: lmind_hotpot_train8000_eval7405_v1_recite_qa_Qwen_Qwen1.5-4B_3e-5_lora2
|
| 12 |
+
results:
|
| 13 |
+
- task:
|
| 14 |
+
name: Causal Language Modeling
|
| 15 |
+
type: text-generation
|
| 16 |
+
dataset:
|
| 17 |
+
name: tyzhu/lmind_hotpot_train8000_eval7405_v1_recite_qa
|
| 18 |
+
type: tyzhu/lmind_hotpot_train8000_eval7405_v1_recite_qa
|
| 19 |
+
metrics:
|
| 20 |
+
- name: Accuracy
|
| 21 |
+
type: accuracy
|
| 22 |
+
value: 0.7263318777292577
|
| 23 |
library_name: peft
|
| 24 |
---
|
| 25 |
|
|
|
|
| 28 |
|
| 29 |
# lmind_hotpot_train8000_eval7405_v1_recite_qa_Qwen_Qwen1.5-4B_3e-5_lora2
|
| 30 |
|
| 31 |
+
This model is a fine-tuned version of [Qwen/Qwen1.5-4B](https://huggingface.co/Qwen/Qwen1.5-4B) on the tyzhu/lmind_hotpot_train8000_eval7405_v1_recite_qa dataset.
|
| 32 |
It achieves the following results on the evaluation set:
|
| 33 |
- Loss: 0.9440
|
| 34 |
- Accuracy: 0.7263
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.995409685563462,
|
| 3 |
+
"eval_accuracy": 0.7263318777292577,
|
| 4 |
+
"eval_loss": 0.9440018534660339,
|
| 5 |
+
"eval_runtime": 8.9976,
|
| 6 |
+
"eval_samples": 500,
|
| 7 |
+
"eval_samples_per_second": 55.571,
|
| 8 |
+
"eval_steps_per_second": 7.002,
|
| 9 |
+
"perplexity": 2.570246615048117,
|
| 10 |
+
"total_flos": 2.2953223726028554e+18,
|
| 11 |
+
"train_loss": 1.0977098983400344,
|
| 12 |
+
"train_runtime": 46950.9086,
|
| 13 |
+
"train_samples": 34854,
|
| 14 |
+
"train_samples_per_second": 14.847,
|
| 15 |
+
"train_steps_per_second": 0.464
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.995409685563462,
|
| 3 |
+
"eval_accuracy": 0.7263318777292577,
|
| 4 |
+
"eval_loss": 0.9440018534660339,
|
| 5 |
+
"eval_runtime": 8.9976,
|
| 6 |
+
"eval_samples": 500,
|
| 7 |
+
"eval_samples_per_second": 55.571,
|
| 8 |
+
"eval_steps_per_second": 7.002,
|
| 9 |
+
"perplexity": 2.570246615048117
|
| 10 |
+
}
|
tokenizer.json
CHANGED
|
@@ -1,11 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"version": "1.0",
|
| 3 |
-
"truncation":
|
| 4 |
-
"direction": "Right",
|
| 5 |
-
"max_length": 1024,
|
| 6 |
-
"strategy": "LongestFirst",
|
| 7 |
-
"stride": 0
|
| 8 |
-
},
|
| 9 |
"padding": null,
|
| 10 |
"added_tokens": [
|
| 11 |
{
|
|
|
|
| 1 |
{
|
| 2 |
"version": "1.0",
|
| 3 |
+
"truncation": null,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
"padding": null,
|
| 5 |
"added_tokens": [
|
| 6 |
{
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.995409685563462,
|
| 3 |
+
"total_flos": 2.2953223726028554e+18,
|
| 4 |
+
"train_loss": 1.0977098983400344,
|
| 5 |
+
"train_runtime": 46950.9086,
|
| 6 |
+
"train_samples": 34854,
|
| 7 |
+
"train_samples_per_second": 14.847,
|
| 8 |
+
"train_steps_per_second": 0.464
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1729 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 19.995409685563462,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 21780,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.09180628873077806,
|
| 13 |
+
"grad_norm": 0.289526104927063,
|
| 14 |
+
"learning_rate": 3e-05,
|
| 15 |
+
"loss": 1.7578,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.18361257746155613,
|
| 20 |
+
"grad_norm": 0.27366071939468384,
|
| 21 |
+
"learning_rate": 3e-05,
|
| 22 |
+
"loss": 1.6025,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.2754188661923342,
|
| 27 |
+
"grad_norm": 0.2666202485561371,
|
| 28 |
+
"learning_rate": 3e-05,
|
| 29 |
+
"loss": 1.5821,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.36722515492311225,
|
| 34 |
+
"grad_norm": 0.25359421968460083,
|
| 35 |
+
"learning_rate": 3e-05,
|
| 36 |
+
"loss": 1.5884,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.4590314436538903,
|
| 41 |
+
"grad_norm": 0.26042038202285767,
|
| 42 |
+
"learning_rate": 3e-05,
|
| 43 |
+
"loss": 1.5873,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.5508377323846684,
|
| 48 |
+
"grad_norm": 0.25545641779899597,
|
| 49 |
+
"learning_rate": 3e-05,
|
| 50 |
+
"loss": 1.5861,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.6426440211154464,
|
| 55 |
+
"grad_norm": 0.2685534358024597,
|
| 56 |
+
"learning_rate": 3e-05,
|
| 57 |
+
"loss": 1.5889,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.7344503098462245,
|
| 62 |
+
"grad_norm": 0.30292272567749023,
|
| 63 |
+
"learning_rate": 3e-05,
|
| 64 |
+
"loss": 1.6001,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.8262565985770025,
|
| 69 |
+
"grad_norm": 0.3182919919490814,
|
| 70 |
+
"learning_rate": 3e-05,
|
| 71 |
+
"loss": 1.5712,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.9180628873077806,
|
| 76 |
+
"grad_norm": 0.31249287724494934,
|
| 77 |
+
"learning_rate": 3e-05,
|
| 78 |
+
"loss": 1.584,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.999770484278173,
|
| 83 |
+
"eval_accuracy": 0.6762416302765648,
|
| 84 |
+
"eval_loss": 1.5102524757385254,
|
| 85 |
+
"eval_runtime": 8.9715,
|
| 86 |
+
"eval_samples_per_second": 55.732,
|
| 87 |
+
"eval_steps_per_second": 7.022,
|
| 88 |
+
"step": 1089
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.0098691760385587,
|
| 92 |
+
"grad_norm": 0.2788257300853729,
|
| 93 |
+
"learning_rate": 3e-05,
|
| 94 |
+
"loss": 1.5741,
|
| 95 |
+
"step": 1100
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.1016754647693368,
|
| 99 |
+
"grad_norm": 0.3435765504837036,
|
| 100 |
+
"learning_rate": 3e-05,
|
| 101 |
+
"loss": 1.5538,
|
| 102 |
+
"step": 1200
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.1934817535001148,
|
| 106 |
+
"grad_norm": 0.42897695302963257,
|
| 107 |
+
"learning_rate": 3e-05,
|
| 108 |
+
"loss": 1.548,
|
| 109 |
+
"step": 1300
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.2852880422308928,
|
| 113 |
+
"grad_norm": 0.3922116160392761,
|
| 114 |
+
"learning_rate": 3e-05,
|
| 115 |
+
"loss": 1.5439,
|
| 116 |
+
"step": 1400
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.377094330961671,
|
| 120 |
+
"grad_norm": 0.39680397510528564,
|
| 121 |
+
"learning_rate": 3e-05,
|
| 122 |
+
"loss": 1.5422,
|
| 123 |
+
"step": 1500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.468900619692449,
|
| 127 |
+
"grad_norm": 0.42543351650238037,
|
| 128 |
+
"learning_rate": 3e-05,
|
| 129 |
+
"loss": 1.5569,
|
| 130 |
+
"step": 1600
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.560706908423227,
|
| 134 |
+
"grad_norm": 0.41228362917900085,
|
| 135 |
+
"learning_rate": 3e-05,
|
| 136 |
+
"loss": 1.5229,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.652513197154005,
|
| 141 |
+
"grad_norm": 0.46888694167137146,
|
| 142 |
+
"learning_rate": 3e-05,
|
| 143 |
+
"loss": 1.5442,
|
| 144 |
+
"step": 1800
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.744319485884783,
|
| 148 |
+
"grad_norm": 0.4511169195175171,
|
| 149 |
+
"learning_rate": 3e-05,
|
| 150 |
+
"loss": 1.538,
|
| 151 |
+
"step": 1900
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.836125774615561,
|
| 155 |
+
"grad_norm": 0.46713030338287354,
|
| 156 |
+
"learning_rate": 3e-05,
|
| 157 |
+
"loss": 1.5343,
|
| 158 |
+
"step": 2000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.9279320633463393,
|
| 162 |
+
"grad_norm": 0.5275651812553406,
|
| 163 |
+
"learning_rate": 3e-05,
|
| 164 |
+
"loss": 1.5504,
|
| 165 |
+
"step": 2100
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.999540968556346,
|
| 169 |
+
"eval_accuracy": 0.6791033478893741,
|
| 170 |
+
"eval_loss": 1.4772261381149292,
|
| 171 |
+
"eval_runtime": 9.0172,
|
| 172 |
+
"eval_samples_per_second": 55.45,
|
| 173 |
+
"eval_steps_per_second": 6.987,
|
| 174 |
+
"step": 2178
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 2.0197383520771175,
|
| 178 |
+
"grad_norm": 0.47441884875297546,
|
| 179 |
+
"learning_rate": 3e-05,
|
| 180 |
+
"loss": 1.5331,
|
| 181 |
+
"step": 2200
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 2.1115446408078955,
|
| 185 |
+
"grad_norm": 0.47345635294914246,
|
| 186 |
+
"learning_rate": 3e-05,
|
| 187 |
+
"loss": 1.4902,
|
| 188 |
+
"step": 2300
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.2033509295386735,
|
| 192 |
+
"grad_norm": 0.592490017414093,
|
| 193 |
+
"learning_rate": 3e-05,
|
| 194 |
+
"loss": 1.4996,
|
| 195 |
+
"step": 2400
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.2951572182694515,
|
| 199 |
+
"grad_norm": 0.5280076265335083,
|
| 200 |
+
"learning_rate": 3e-05,
|
| 201 |
+
"loss": 1.4949,
|
| 202 |
+
"step": 2500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 2.3869635070002295,
|
| 206 |
+
"grad_norm": 0.5791444182395935,
|
| 207 |
+
"learning_rate": 3e-05,
|
| 208 |
+
"loss": 1.499,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.4787697957310075,
|
| 213 |
+
"grad_norm": 0.5848264098167419,
|
| 214 |
+
"learning_rate": 3e-05,
|
| 215 |
+
"loss": 1.479,
|
| 216 |
+
"step": 2700
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.5705760844617855,
|
| 220 |
+
"grad_norm": 0.5598397254943848,
|
| 221 |
+
"learning_rate": 3e-05,
|
| 222 |
+
"loss": 1.4938,
|
| 223 |
+
"step": 2800
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 2.6623823731925635,
|
| 227 |
+
"grad_norm": 0.5774019360542297,
|
| 228 |
+
"learning_rate": 3e-05,
|
| 229 |
+
"loss": 1.4884,
|
| 230 |
+
"step": 2900
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.754188661923342,
|
| 234 |
+
"grad_norm": 0.6278976202011108,
|
| 235 |
+
"learning_rate": 3e-05,
|
| 236 |
+
"loss": 1.5027,
|
| 237 |
+
"step": 3000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.84599495065412,
|
| 241 |
+
"grad_norm": 0.5700748562812805,
|
| 242 |
+
"learning_rate": 3e-05,
|
| 243 |
+
"loss": 1.4777,
|
| 244 |
+
"step": 3100
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 2.937801239384898,
|
| 248 |
+
"grad_norm": 0.6338950395584106,
|
| 249 |
+
"learning_rate": 3e-05,
|
| 250 |
+
"loss": 1.4842,
|
| 251 |
+
"step": 3200
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 2.9993114528345193,
|
| 255 |
+
"eval_accuracy": 0.6811615720524018,
|
| 256 |
+
"eval_loss": 1.4502207040786743,
|
| 257 |
+
"eval_runtime": 9.3503,
|
| 258 |
+
"eval_samples_per_second": 53.474,
|
| 259 |
+
"eval_steps_per_second": 6.738,
|
| 260 |
+
"step": 3267
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 3.029607528115676,
|
| 264 |
+
"grad_norm": 0.6205599904060364,
|
| 265 |
+
"learning_rate": 3e-05,
|
| 266 |
+
"loss": 1.4695,
|
| 267 |
+
"step": 3300
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 3.121413816846454,
|
| 271 |
+
"grad_norm": 0.6470921635627747,
|
| 272 |
+
"learning_rate": 3e-05,
|
| 273 |
+
"loss": 1.4401,
|
| 274 |
+
"step": 3400
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 3.213220105577232,
|
| 278 |
+
"grad_norm": 0.8113517761230469,
|
| 279 |
+
"learning_rate": 3e-05,
|
| 280 |
+
"loss": 1.4312,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.30502639430801,
|
| 285 |
+
"grad_norm": 0.6976670026779175,
|
| 286 |
+
"learning_rate": 3e-05,
|
| 287 |
+
"loss": 1.4548,
|
| 288 |
+
"step": 3600
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.396832683038788,
|
| 292 |
+
"grad_norm": 0.7569802403450012,
|
| 293 |
+
"learning_rate": 3e-05,
|
| 294 |
+
"loss": 1.4447,
|
| 295 |
+
"step": 3700
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 3.488638971769566,
|
| 299 |
+
"grad_norm": 0.8047822117805481,
|
| 300 |
+
"learning_rate": 3e-05,
|
| 301 |
+
"loss": 1.434,
|
| 302 |
+
"step": 3800
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.580445260500344,
|
| 306 |
+
"grad_norm": 0.710166871547699,
|
| 307 |
+
"learning_rate": 3e-05,
|
| 308 |
+
"loss": 1.4283,
|
| 309 |
+
"step": 3900
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.672251549231122,
|
| 313 |
+
"grad_norm": 0.7864311933517456,
|
| 314 |
+
"learning_rate": 3e-05,
|
| 315 |
+
"loss": 1.4493,
|
| 316 |
+
"step": 4000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.7640578379619005,
|
| 320 |
+
"grad_norm": 0.7331141829490662,
|
| 321 |
+
"learning_rate": 3e-05,
|
| 322 |
+
"loss": 1.4259,
|
| 323 |
+
"step": 4100
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.8558641266926785,
|
| 327 |
+
"grad_norm": 0.7041341662406921,
|
| 328 |
+
"learning_rate": 3e-05,
|
| 329 |
+
"loss": 1.4316,
|
| 330 |
+
"step": 4200
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.9476704154234565,
|
| 334 |
+
"grad_norm": 0.6956498622894287,
|
| 335 |
+
"learning_rate": 3e-05,
|
| 336 |
+
"loss": 1.427,
|
| 337 |
+
"step": 4300
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 4.0,
|
| 341 |
+
"eval_accuracy": 0.6837205240174673,
|
| 342 |
+
"eval_loss": 1.4199937582015991,
|
| 343 |
+
"eval_runtime": 9.1672,
|
| 344 |
+
"eval_samples_per_second": 54.542,
|
| 345 |
+
"eval_steps_per_second": 6.872,
|
| 346 |
+
"step": 4357
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 4.039476704154235,
|
| 350 |
+
"grad_norm": 0.7389739751815796,
|
| 351 |
+
"learning_rate": 3e-05,
|
| 352 |
+
"loss": 1.4099,
|
| 353 |
+
"step": 4400
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 4.131282992885013,
|
| 357 |
+
"grad_norm": 0.7883840203285217,
|
| 358 |
+
"learning_rate": 3e-05,
|
| 359 |
+
"loss": 1.3709,
|
| 360 |
+
"step": 4500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 4.223089281615791,
|
| 364 |
+
"grad_norm": 0.7341814041137695,
|
| 365 |
+
"learning_rate": 3e-05,
|
| 366 |
+
"loss": 1.3743,
|
| 367 |
+
"step": 4600
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 4.314895570346569,
|
| 371 |
+
"grad_norm": 0.9007183909416199,
|
| 372 |
+
"learning_rate": 3e-05,
|
| 373 |
+
"loss": 1.3929,
|
| 374 |
+
"step": 4700
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 4.406701859077347,
|
| 378 |
+
"grad_norm": 0.8208268284797668,
|
| 379 |
+
"learning_rate": 3e-05,
|
| 380 |
+
"loss": 1.3843,
|
| 381 |
+
"step": 4800
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 4.498508147808125,
|
| 385 |
+
"grad_norm": 0.7786016464233398,
|
| 386 |
+
"learning_rate": 3e-05,
|
| 387 |
+
"loss": 1.3789,
|
| 388 |
+
"step": 4900
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 4.590314436538903,
|
| 392 |
+
"grad_norm": 0.9414695501327515,
|
| 393 |
+
"learning_rate": 3e-05,
|
| 394 |
+
"loss": 1.3812,
|
| 395 |
+
"step": 5000
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 4.682120725269681,
|
| 399 |
+
"grad_norm": 0.7854552268981934,
|
| 400 |
+
"learning_rate": 3e-05,
|
| 401 |
+
"loss": 1.3851,
|
| 402 |
+
"step": 5100
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 4.773927014000459,
|
| 406 |
+
"grad_norm": 0.8319596648216248,
|
| 407 |
+
"learning_rate": 3e-05,
|
| 408 |
+
"loss": 1.3846,
|
| 409 |
+
"step": 5200
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 4.865733302731237,
|
| 413 |
+
"grad_norm": 0.8832118511199951,
|
| 414 |
+
"learning_rate": 3e-05,
|
| 415 |
+
"loss": 1.3845,
|
| 416 |
+
"step": 5300
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 4.957539591462015,
|
| 420 |
+
"grad_norm": 0.8607555627822876,
|
| 421 |
+
"learning_rate": 3e-05,
|
| 422 |
+
"loss": 1.3827,
|
| 423 |
+
"step": 5400
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.999770484278173,
|
| 427 |
+
"eval_accuracy": 0.6860465793304221,
|
| 428 |
+
"eval_loss": 1.3910759687423706,
|
| 429 |
+
"eval_runtime": 8.9399,
|
| 430 |
+
"eval_samples_per_second": 55.929,
|
| 431 |
+
"eval_steps_per_second": 7.047,
|
| 432 |
+
"step": 5446
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 5.049345880192793,
|
| 436 |
+
"grad_norm": 0.8777875900268555,
|
| 437 |
+
"learning_rate": 3e-05,
|
| 438 |
+
"loss": 1.3512,
|
| 439 |
+
"step": 5500
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 5.141152168923571,
|
| 443 |
+
"grad_norm": 0.9193658232688904,
|
| 444 |
+
"learning_rate": 3e-05,
|
| 445 |
+
"loss": 1.3124,
|
| 446 |
+
"step": 5600
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 5.232958457654349,
|
| 450 |
+
"grad_norm": 0.9822832345962524,
|
| 451 |
+
"learning_rate": 3e-05,
|
| 452 |
+
"loss": 1.3189,
|
| 453 |
+
"step": 5700
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 5.324764746385127,
|
| 457 |
+
"grad_norm": 0.9231218099594116,
|
| 458 |
+
"learning_rate": 3e-05,
|
| 459 |
+
"loss": 1.3321,
|
| 460 |
+
"step": 5800
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 5.416571035115905,
|
| 464 |
+
"grad_norm": 0.961618185043335,
|
| 465 |
+
"learning_rate": 3e-05,
|
| 466 |
+
"loss": 1.3275,
|
| 467 |
+
"step": 5900
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 5.508377323846684,
|
| 471 |
+
"grad_norm": 1.1759928464889526,
|
| 472 |
+
"learning_rate": 3e-05,
|
| 473 |
+
"loss": 1.3301,
|
| 474 |
+
"step": 6000
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 5.600183612577462,
|
| 478 |
+
"grad_norm": 1.0055111646652222,
|
| 479 |
+
"learning_rate": 3e-05,
|
| 480 |
+
"loss": 1.3261,
|
| 481 |
+
"step": 6100
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 5.69198990130824,
|
| 485 |
+
"grad_norm": 0.9605348110198975,
|
| 486 |
+
"learning_rate": 3e-05,
|
| 487 |
+
"loss": 1.3329,
|
| 488 |
+
"step": 6200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 5.783796190039018,
|
| 492 |
+
"grad_norm": 1.0969476699829102,
|
| 493 |
+
"learning_rate": 3e-05,
|
| 494 |
+
"loss": 1.3347,
|
| 495 |
+
"step": 6300
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 5.875602478769796,
|
| 499 |
+
"grad_norm": 0.9841852188110352,
|
| 500 |
+
"learning_rate": 3e-05,
|
| 501 |
+
"loss": 1.3215,
|
| 502 |
+
"step": 6400
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 5.967408767500574,
|
| 506 |
+
"grad_norm": 1.0173933506011963,
|
| 507 |
+
"learning_rate": 3e-05,
|
| 508 |
+
"loss": 1.3425,
|
| 509 |
+
"step": 6500
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 5.999540968556346,
|
| 513 |
+
"eval_accuracy": 0.6886841339155749,
|
| 514 |
+
"eval_loss": 1.3614963293075562,
|
| 515 |
+
"eval_runtime": 8.9983,
|
| 516 |
+
"eval_samples_per_second": 55.566,
|
| 517 |
+
"eval_steps_per_second": 7.001,
|
| 518 |
+
"step": 6535
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 6.059215056231352,
|
| 522 |
+
"grad_norm": 0.9646373987197876,
|
| 523 |
+
"learning_rate": 3e-05,
|
| 524 |
+
"loss": 1.287,
|
| 525 |
+
"step": 6600
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 6.15102134496213,
|
| 529 |
+
"grad_norm": 0.8929613828659058,
|
| 530 |
+
"learning_rate": 3e-05,
|
| 531 |
+
"loss": 1.2574,
|
| 532 |
+
"step": 6700
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 6.242827633692908,
|
| 536 |
+
"grad_norm": 1.285346508026123,
|
| 537 |
+
"learning_rate": 3e-05,
|
| 538 |
+
"loss": 1.2762,
|
| 539 |
+
"step": 6800
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 6.334633922423686,
|
| 543 |
+
"grad_norm": 1.102123498916626,
|
| 544 |
+
"learning_rate": 3e-05,
|
| 545 |
+
"loss": 1.2667,
|
| 546 |
+
"step": 6900
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 6.426440211154464,
|
| 550 |
+
"grad_norm": 1.021745204925537,
|
| 551 |
+
"learning_rate": 3e-05,
|
| 552 |
+
"loss": 1.2698,
|
| 553 |
+
"step": 7000
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 6.518246499885242,
|
| 557 |
+
"grad_norm": 1.1759482622146606,
|
| 558 |
+
"learning_rate": 3e-05,
|
| 559 |
+
"loss": 1.2781,
|
| 560 |
+
"step": 7100
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 6.61005278861602,
|
| 564 |
+
"grad_norm": 1.2193723917007446,
|
| 565 |
+
"learning_rate": 3e-05,
|
| 566 |
+
"loss": 1.2784,
|
| 567 |
+
"step": 7200
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 6.701859077346798,
|
| 571 |
+
"grad_norm": 1.1053309440612793,
|
| 572 |
+
"learning_rate": 3e-05,
|
| 573 |
+
"loss": 1.2732,
|
| 574 |
+
"step": 7300
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 6.793665366077576,
|
| 578 |
+
"grad_norm": 1.7023396492004395,
|
| 579 |
+
"learning_rate": 3e-05,
|
| 580 |
+
"loss": 1.269,
|
| 581 |
+
"step": 7400
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 6.885471654808354,
|
| 585 |
+
"grad_norm": 1.0934760570526123,
|
| 586 |
+
"learning_rate": 3e-05,
|
| 587 |
+
"loss": 1.2691,
|
| 588 |
+
"step": 7500
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 6.977277943539132,
|
| 592 |
+
"grad_norm": 1.0586143732070923,
|
| 593 |
+
"learning_rate": 3e-05,
|
| 594 |
+
"loss": 1.2738,
|
| 595 |
+
"step": 7600
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 6.999311452834519,
|
| 599 |
+
"eval_accuracy": 0.6910072780203784,
|
| 600 |
+
"eval_loss": 1.3299835920333862,
|
| 601 |
+
"eval_runtime": 9.0186,
|
| 602 |
+
"eval_samples_per_second": 55.441,
|
| 603 |
+
"eval_steps_per_second": 6.986,
|
| 604 |
+
"step": 7624
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 7.06908423226991,
|
| 608 |
+
"grad_norm": 1.2295095920562744,
|
| 609 |
+
"learning_rate": 3e-05,
|
| 610 |
+
"loss": 1.2221,
|
| 611 |
+
"step": 7700
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 7.160890521000688,
|
| 615 |
+
"grad_norm": 1.0440340042114258,
|
| 616 |
+
"learning_rate": 3e-05,
|
| 617 |
+
"loss": 1.2119,
|
| 618 |
+
"step": 7800
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 7.252696809731467,
|
| 622 |
+
"grad_norm": 1.1318169832229614,
|
| 623 |
+
"learning_rate": 3e-05,
|
| 624 |
+
"loss": 1.2061,
|
| 625 |
+
"step": 7900
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 7.344503098462245,
|
| 629 |
+
"grad_norm": 1.2543174028396606,
|
| 630 |
+
"learning_rate": 3e-05,
|
| 631 |
+
"loss": 1.2222,
|
| 632 |
+
"step": 8000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 7.436309387193023,
|
| 636 |
+
"grad_norm": 1.098528504371643,
|
| 637 |
+
"learning_rate": 3e-05,
|
| 638 |
+
"loss": 1.211,
|
| 639 |
+
"step": 8100
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 7.528115675923801,
|
| 643 |
+
"grad_norm": 1.5505329370498657,
|
| 644 |
+
"learning_rate": 3e-05,
|
| 645 |
+
"loss": 1.2313,
|
| 646 |
+
"step": 8200
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 7.619921964654579,
|
| 650 |
+
"grad_norm": 1.2159889936447144,
|
| 651 |
+
"learning_rate": 3e-05,
|
| 652 |
+
"loss": 1.2155,
|
| 653 |
+
"step": 8300
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 7.711728253385357,
|
| 657 |
+
"grad_norm": 1.2545368671417236,
|
| 658 |
+
"learning_rate": 3e-05,
|
| 659 |
+
"loss": 1.2125,
|
| 660 |
+
"step": 8400
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 7.803534542116135,
|
| 664 |
+
"grad_norm": 1.0893586874008179,
|
| 665 |
+
"learning_rate": 3e-05,
|
| 666 |
+
"loss": 1.2067,
|
| 667 |
+
"step": 8500
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 7.895340830846913,
|
| 671 |
+
"grad_norm": 1.2962942123413086,
|
| 672 |
+
"learning_rate": 3e-05,
|
| 673 |
+
"loss": 1.2123,
|
| 674 |
+
"step": 8600
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 7.987147119577691,
|
| 678 |
+
"grad_norm": 1.0764884948730469,
|
| 679 |
+
"learning_rate": 3e-05,
|
| 680 |
+
"loss": 1.2283,
|
| 681 |
+
"step": 8700
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 8.0,
|
| 685 |
+
"eval_accuracy": 0.6934876273653566,
|
| 686 |
+
"eval_loss": 1.3020232915878296,
|
| 687 |
+
"eval_runtime": 8.9433,
|
| 688 |
+
"eval_samples_per_second": 55.908,
|
| 689 |
+
"eval_steps_per_second": 7.044,
|
| 690 |
+
"step": 8714
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 8.07895340830847,
|
| 694 |
+
"grad_norm": 1.2428829669952393,
|
| 695 |
+
"learning_rate": 3e-05,
|
| 696 |
+
"loss": 1.1629,
|
| 697 |
+
"step": 8800
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 8.170759697039248,
|
| 701 |
+
"grad_norm": 1.19692862033844,
|
| 702 |
+
"learning_rate": 3e-05,
|
| 703 |
+
"loss": 1.1448,
|
| 704 |
+
"step": 8900
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 8.262565985770026,
|
| 708 |
+
"grad_norm": 1.3424954414367676,
|
| 709 |
+
"learning_rate": 3e-05,
|
| 710 |
+
"loss": 1.1617,
|
| 711 |
+
"step": 9000
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 8.354372274500804,
|
| 715 |
+
"grad_norm": 1.3499901294708252,
|
| 716 |
+
"learning_rate": 3e-05,
|
| 717 |
+
"loss": 1.1564,
|
| 718 |
+
"step": 9100
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 8.446178563231582,
|
| 722 |
+
"grad_norm": 1.183600664138794,
|
| 723 |
+
"learning_rate": 3e-05,
|
| 724 |
+
"loss": 1.1524,
|
| 725 |
+
"step": 9200
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 8.53798485196236,
|
| 729 |
+
"grad_norm": 1.3151459693908691,
|
| 730 |
+
"learning_rate": 3e-05,
|
| 731 |
+
"loss": 1.1499,
|
| 732 |
+
"step": 9300
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 8.629791140693138,
|
| 736 |
+
"grad_norm": 1.3484901189804077,
|
| 737 |
+
"learning_rate": 3e-05,
|
| 738 |
+
"loss": 1.1716,
|
| 739 |
+
"step": 9400
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 8.721597429423916,
|
| 743 |
+
"grad_norm": 1.326663851737976,
|
| 744 |
+
"learning_rate": 3e-05,
|
| 745 |
+
"loss": 1.1665,
|
| 746 |
+
"step": 9500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 8.813403718154694,
|
| 750 |
+
"grad_norm": 1.315091609954834,
|
| 751 |
+
"learning_rate": 3e-05,
|
| 752 |
+
"loss": 1.1713,
|
| 753 |
+
"step": 9600
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 8.905210006885472,
|
| 757 |
+
"grad_norm": 1.3116127252578735,
|
| 758 |
+
"learning_rate": 3e-05,
|
| 759 |
+
"loss": 1.1669,
|
| 760 |
+
"step": 9700
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 8.99701629561625,
|
| 764 |
+
"grad_norm": 1.4249849319458008,
|
| 765 |
+
"learning_rate": 3e-05,
|
| 766 |
+
"loss": 1.1788,
|
| 767 |
+
"step": 9800
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 8.999770484278173,
|
| 771 |
+
"eval_accuracy": 0.6963406113537118,
|
| 772 |
+
"eval_loss": 1.272213339805603,
|
| 773 |
+
"eval_runtime": 8.9482,
|
| 774 |
+
"eval_samples_per_second": 55.877,
|
| 775 |
+
"eval_steps_per_second": 7.041,
|
| 776 |
+
"step": 9803
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 9.088822584347028,
|
| 780 |
+
"grad_norm": 1.412617802619934,
|
| 781 |
+
"learning_rate": 3e-05,
|
| 782 |
+
"loss": 1.0912,
|
| 783 |
+
"step": 9900
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 9.180628873077806,
|
| 787 |
+
"grad_norm": 1.2806413173675537,
|
| 788 |
+
"learning_rate": 3e-05,
|
| 789 |
+
"loss": 1.1013,
|
| 790 |
+
"step": 10000
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 9.272435161808584,
|
| 794 |
+
"grad_norm": 1.8053137063980103,
|
| 795 |
+
"learning_rate": 3e-05,
|
| 796 |
+
"loss": 1.1071,
|
| 797 |
+
"step": 10100
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 9.364241450539362,
|
| 801 |
+
"grad_norm": 1.352771520614624,
|
| 802 |
+
"learning_rate": 3e-05,
|
| 803 |
+
"loss": 1.1043,
|
| 804 |
+
"step": 10200
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 9.45604773927014,
|
| 808 |
+
"grad_norm": 1.5698919296264648,
|
| 809 |
+
"learning_rate": 3e-05,
|
| 810 |
+
"loss": 1.1187,
|
| 811 |
+
"step": 10300
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 9.547854028000918,
|
| 815 |
+
"grad_norm": 1.4473572969436646,
|
| 816 |
+
"learning_rate": 3e-05,
|
| 817 |
+
"loss": 1.0991,
|
| 818 |
+
"step": 10400
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 9.639660316731696,
|
| 822 |
+
"grad_norm": 1.5458990335464478,
|
| 823 |
+
"learning_rate": 3e-05,
|
| 824 |
+
"loss": 1.1168,
|
| 825 |
+
"step": 10500
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 9.731466605462474,
|
| 829 |
+
"grad_norm": 1.3577615022659302,
|
| 830 |
+
"learning_rate": 3e-05,
|
| 831 |
+
"loss": 1.1081,
|
| 832 |
+
"step": 10600
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 9.823272894193252,
|
| 836 |
+
"grad_norm": 1.6055794954299927,
|
| 837 |
+
"learning_rate": 3e-05,
|
| 838 |
+
"loss": 1.1117,
|
| 839 |
+
"step": 10700
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 9.91507918292403,
|
| 843 |
+
"grad_norm": 1.5611170530319214,
|
| 844 |
+
"learning_rate": 3e-05,
|
| 845 |
+
"loss": 1.1156,
|
| 846 |
+
"step": 10800
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 9.999540968556346,
|
| 850 |
+
"eval_accuracy": 0.6990072780203784,
|
| 851 |
+
"eval_loss": 1.2414618730545044,
|
| 852 |
+
"eval_runtime": 9.5021,
|
| 853 |
+
"eval_samples_per_second": 52.62,
|
| 854 |
+
"eval_steps_per_second": 6.63,
|
| 855 |
+
"step": 10892
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 10.006885471654808,
|
| 859 |
+
"grad_norm": 1.4584790468215942,
|
| 860 |
+
"learning_rate": 3e-05,
|
| 861 |
+
"loss": 1.1099,
|
| 862 |
+
"step": 10900
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 10.098691760385586,
|
| 866 |
+
"grad_norm": 1.4327212572097778,
|
| 867 |
+
"learning_rate": 3e-05,
|
| 868 |
+
"loss": 1.039,
|
| 869 |
+
"step": 11000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 10.190498049116364,
|
| 873 |
+
"grad_norm": 1.4160873889923096,
|
| 874 |
+
"learning_rate": 3e-05,
|
| 875 |
+
"loss": 1.0578,
|
| 876 |
+
"step": 11100
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 10.282304337847142,
|
| 880 |
+
"grad_norm": 1.506165862083435,
|
| 881 |
+
"learning_rate": 3e-05,
|
| 882 |
+
"loss": 1.0481,
|
| 883 |
+
"step": 11200
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 10.37411062657792,
|
| 887 |
+
"grad_norm": 1.6476013660430908,
|
| 888 |
+
"learning_rate": 3e-05,
|
| 889 |
+
"loss": 1.0582,
|
| 890 |
+
"step": 11300
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 10.465916915308698,
|
| 894 |
+
"grad_norm": 1.4203314781188965,
|
| 895 |
+
"learning_rate": 3e-05,
|
| 896 |
+
"loss": 1.0615,
|
| 897 |
+
"step": 11400
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 10.557723204039476,
|
| 901 |
+
"grad_norm": 1.591191053390503,
|
| 902 |
+
"learning_rate": 3e-05,
|
| 903 |
+
"loss": 1.0654,
|
| 904 |
+
"step": 11500
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 10.649529492770254,
|
| 908 |
+
"grad_norm": 1.552139401435852,
|
| 909 |
+
"learning_rate": 3e-05,
|
| 910 |
+
"loss": 1.043,
|
| 911 |
+
"step": 11600
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 10.741335781501032,
|
| 915 |
+
"grad_norm": 1.5005476474761963,
|
| 916 |
+
"learning_rate": 3e-05,
|
| 917 |
+
"loss": 1.0518,
|
| 918 |
+
"step": 11700
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 10.83314207023181,
|
| 922 |
+
"grad_norm": 1.6541969776153564,
|
| 923 |
+
"learning_rate": 3e-05,
|
| 924 |
+
"loss": 1.0617,
|
| 925 |
+
"step": 11800
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 10.924948358962588,
|
| 929 |
+
"grad_norm": 1.4498178958892822,
|
| 930 |
+
"learning_rate": 3e-05,
|
| 931 |
+
"loss": 1.0526,
|
| 932 |
+
"step": 11900
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 10.999311452834519,
|
| 936 |
+
"eval_accuracy": 0.7014876273653566,
|
| 937 |
+
"eval_loss": 1.2131479978561401,
|
| 938 |
+
"eval_runtime": 9.02,
|
| 939 |
+
"eval_samples_per_second": 55.432,
|
| 940 |
+
"eval_steps_per_second": 6.984,
|
| 941 |
+
"step": 11981
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 11.016754647693366,
|
| 945 |
+
"grad_norm": 1.384069561958313,
|
| 946 |
+
"learning_rate": 3e-05,
|
| 947 |
+
"loss": 1.0554,
|
| 948 |
+
"step": 12000
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 11.108560936424144,
|
| 952 |
+
"grad_norm": 1.4845672845840454,
|
| 953 |
+
"learning_rate": 3e-05,
|
| 954 |
+
"loss": 0.9886,
|
| 955 |
+
"step": 12100
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 11.200367225154924,
|
| 959 |
+
"grad_norm": 1.7744626998901367,
|
| 960 |
+
"learning_rate": 3e-05,
|
| 961 |
+
"loss": 0.9969,
|
| 962 |
+
"step": 12200
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 11.292173513885702,
|
| 966 |
+
"grad_norm": 1.6337647438049316,
|
| 967 |
+
"learning_rate": 3e-05,
|
| 968 |
+
"loss": 0.9899,
|
| 969 |
+
"step": 12300
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 11.38397980261648,
|
| 973 |
+
"grad_norm": 2.003005266189575,
|
| 974 |
+
"learning_rate": 3e-05,
|
| 975 |
+
"loss": 1.0111,
|
| 976 |
+
"step": 12400
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 11.475786091347258,
|
| 980 |
+
"grad_norm": 1.968371033668518,
|
| 981 |
+
"learning_rate": 3e-05,
|
| 982 |
+
"loss": 1.0012,
|
| 983 |
+
"step": 12500
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 11.567592380078036,
|
| 987 |
+
"grad_norm": 1.6538879871368408,
|
| 988 |
+
"learning_rate": 3e-05,
|
| 989 |
+
"loss": 0.9972,
|
| 990 |
+
"step": 12600
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 11.659398668808814,
|
| 994 |
+
"grad_norm": 1.6392265558242798,
|
| 995 |
+
"learning_rate": 3e-05,
|
| 996 |
+
"loss": 1.0084,
|
| 997 |
+
"step": 12700
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 11.751204957539592,
|
| 1001 |
+
"grad_norm": 1.7361793518066406,
|
| 1002 |
+
"learning_rate": 3e-05,
|
| 1003 |
+
"loss": 1.0109,
|
| 1004 |
+
"step": 12800
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 11.84301124627037,
|
| 1008 |
+
"grad_norm": 1.4300850629806519,
|
| 1009 |
+
"learning_rate": 3e-05,
|
| 1010 |
+
"loss": 1.0163,
|
| 1011 |
+
"step": 12900
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 11.934817535001148,
|
| 1015 |
+
"grad_norm": 1.6984518766403198,
|
| 1016 |
+
"learning_rate": 3e-05,
|
| 1017 |
+
"loss": 1.0146,
|
| 1018 |
+
"step": 13000
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 12.0,
|
| 1022 |
+
"eval_accuracy": 0.7045123726346434,
|
| 1023 |
+
"eval_loss": 1.1803950071334839,
|
| 1024 |
+
"eval_runtime": 8.9699,
|
| 1025 |
+
"eval_samples_per_second": 55.742,
|
| 1026 |
+
"eval_steps_per_second": 7.024,
|
| 1027 |
+
"step": 13071
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 12.026623823731926,
|
| 1031 |
+
"grad_norm": 1.52531898021698,
|
| 1032 |
+
"learning_rate": 3e-05,
|
| 1033 |
+
"loss": 0.9803,
|
| 1034 |
+
"step": 13100
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 12.118430112462704,
|
| 1038 |
+
"grad_norm": 1.8437868356704712,
|
| 1039 |
+
"learning_rate": 3e-05,
|
| 1040 |
+
"loss": 0.9362,
|
| 1041 |
+
"step": 13200
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 12.210236401193482,
|
| 1045 |
+
"grad_norm": 1.7236285209655762,
|
| 1046 |
+
"learning_rate": 3e-05,
|
| 1047 |
+
"loss": 0.9476,
|
| 1048 |
+
"step": 13300
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 12.30204268992426,
|
| 1052 |
+
"grad_norm": 1.7923431396484375,
|
| 1053 |
+
"learning_rate": 3e-05,
|
| 1054 |
+
"loss": 0.9473,
|
| 1055 |
+
"step": 13400
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 12.393848978655038,
|
| 1059 |
+
"grad_norm": 1.9459409713745117,
|
| 1060 |
+
"learning_rate": 3e-05,
|
| 1061 |
+
"loss": 0.9521,
|
| 1062 |
+
"step": 13500
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 12.485655267385816,
|
| 1066 |
+
"grad_norm": 1.8831307888031006,
|
| 1067 |
+
"learning_rate": 3e-05,
|
| 1068 |
+
"loss": 0.942,
|
| 1069 |
+
"step": 13600
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 12.577461556116594,
|
| 1073 |
+
"grad_norm": 1.629230260848999,
|
| 1074 |
+
"learning_rate": 3e-05,
|
| 1075 |
+
"loss": 0.9558,
|
| 1076 |
+
"step": 13700
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 12.669267844847372,
|
| 1080 |
+
"grad_norm": 1.5318315029144287,
|
| 1081 |
+
"learning_rate": 3e-05,
|
| 1082 |
+
"loss": 0.9525,
|
| 1083 |
+
"step": 13800
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 12.76107413357815,
|
| 1087 |
+
"grad_norm": 1.611336588859558,
|
| 1088 |
+
"learning_rate": 3e-05,
|
| 1089 |
+
"loss": 0.9619,
|
| 1090 |
+
"step": 13900
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 12.852880422308928,
|
| 1094 |
+
"grad_norm": 1.6721709966659546,
|
| 1095 |
+
"learning_rate": 3e-05,
|
| 1096 |
+
"loss": 0.9619,
|
| 1097 |
+
"step": 14000
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 12.944686711039706,
|
| 1101 |
+
"grad_norm": 1.8074623346328735,
|
| 1102 |
+
"learning_rate": 3e-05,
|
| 1103 |
+
"loss": 0.9613,
|
| 1104 |
+
"step": 14100
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 12.999770484278173,
|
| 1108 |
+
"eval_accuracy": 0.7071382823871907,
|
| 1109 |
+
"eval_loss": 1.1507638692855835,
|
| 1110 |
+
"eval_runtime": 8.9983,
|
| 1111 |
+
"eval_samples_per_second": 55.566,
|
| 1112 |
+
"eval_steps_per_second": 7.001,
|
| 1113 |
+
"step": 14160
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 13.036492999770484,
|
| 1117 |
+
"grad_norm": 2.195594549179077,
|
| 1118 |
+
"learning_rate": 3e-05,
|
| 1119 |
+
"loss": 0.9259,
|
| 1120 |
+
"step": 14200
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 13.128299288501262,
|
| 1124 |
+
"grad_norm": 1.8173458576202393,
|
| 1125 |
+
"learning_rate": 3e-05,
|
| 1126 |
+
"loss": 0.8902,
|
| 1127 |
+
"step": 14300
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 13.22010557723204,
|
| 1131 |
+
"grad_norm": 1.7481939792633057,
|
| 1132 |
+
"learning_rate": 3e-05,
|
| 1133 |
+
"loss": 0.8859,
|
| 1134 |
+
"step": 14400
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 13.311911865962818,
|
| 1138 |
+
"grad_norm": 1.938438892364502,
|
| 1139 |
+
"learning_rate": 3e-05,
|
| 1140 |
+
"loss": 0.899,
|
| 1141 |
+
"step": 14500
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 13.403718154693596,
|
| 1145 |
+
"grad_norm": 1.8565011024475098,
|
| 1146 |
+
"learning_rate": 3e-05,
|
| 1147 |
+
"loss": 0.889,
|
| 1148 |
+
"step": 14600
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 13.495524443424374,
|
| 1152 |
+
"grad_norm": 1.6509944200515747,
|
| 1153 |
+
"learning_rate": 3e-05,
|
| 1154 |
+
"loss": 0.9182,
|
| 1155 |
+
"step": 14700
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 13.587330732155152,
|
| 1159 |
+
"grad_norm": 1.9225726127624512,
|
| 1160 |
+
"learning_rate": 3e-05,
|
| 1161 |
+
"loss": 0.9091,
|
| 1162 |
+
"step": 14800
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 13.67913702088593,
|
| 1166 |
+
"grad_norm": 1.7917280197143555,
|
| 1167 |
+
"learning_rate": 3e-05,
|
| 1168 |
+
"loss": 0.9137,
|
| 1169 |
+
"step": 14900
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 13.770943309616708,
|
| 1173 |
+
"grad_norm": 2.0736453533172607,
|
| 1174 |
+
"learning_rate": 3e-05,
|
| 1175 |
+
"loss": 0.904,
|
| 1176 |
+
"step": 15000
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 13.862749598347486,
|
| 1180 |
+
"grad_norm": 2.1191747188568115,
|
| 1181 |
+
"learning_rate": 3e-05,
|
| 1182 |
+
"loss": 0.9146,
|
| 1183 |
+
"step": 15100
|
| 1184 |
+
},
|
| 1185 |
+
{
|
| 1186 |
+
"epoch": 13.954555887078264,
|
| 1187 |
+
"grad_norm": 1.8331027030944824,
|
| 1188 |
+
"learning_rate": 3e-05,
|
| 1189 |
+
"loss": 0.9109,
|
| 1190 |
+
"step": 15200
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 13.999540968556346,
|
| 1194 |
+
"eval_accuracy": 0.7097409024745269,
|
| 1195 |
+
"eval_loss": 1.12144935131073,
|
| 1196 |
+
"eval_runtime": 8.9899,
|
| 1197 |
+
"eval_samples_per_second": 55.618,
|
| 1198 |
+
"eval_steps_per_second": 7.008,
|
| 1199 |
+
"step": 15249
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 14.046362175809042,
|
| 1203 |
+
"grad_norm": 1.818524718284607,
|
| 1204 |
+
"learning_rate": 3e-05,
|
| 1205 |
+
"loss": 0.8787,
|
| 1206 |
+
"step": 15300
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 14.13816846453982,
|
| 1210 |
+
"grad_norm": 1.9324177503585815,
|
| 1211 |
+
"learning_rate": 3e-05,
|
| 1212 |
+
"loss": 0.8487,
|
| 1213 |
+
"step": 15400
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 14.229974753270598,
|
| 1217 |
+
"grad_norm": 1.952480435371399,
|
| 1218 |
+
"learning_rate": 3e-05,
|
| 1219 |
+
"loss": 0.8526,
|
| 1220 |
+
"step": 15500
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 14.321781042001376,
|
| 1224 |
+
"grad_norm": 1.9058892726898193,
|
| 1225 |
+
"learning_rate": 3e-05,
|
| 1226 |
+
"loss": 0.8433,
|
| 1227 |
+
"step": 15600
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 14.413587330732156,
|
| 1231 |
+
"grad_norm": 1.9198521375656128,
|
| 1232 |
+
"learning_rate": 3e-05,
|
| 1233 |
+
"loss": 0.8591,
|
| 1234 |
+
"step": 15700
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 14.505393619462934,
|
| 1238 |
+
"grad_norm": 2.374208927154541,
|
| 1239 |
+
"learning_rate": 3e-05,
|
| 1240 |
+
"loss": 0.8643,
|
| 1241 |
+
"step": 15800
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 14.597199908193712,
|
| 1245 |
+
"grad_norm": 1.8864604234695435,
|
| 1246 |
+
"learning_rate": 3e-05,
|
| 1247 |
+
"loss": 0.8623,
|
| 1248 |
+
"step": 15900
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 14.68900619692449,
|
| 1252 |
+
"grad_norm": 1.9877722263336182,
|
| 1253 |
+
"learning_rate": 3e-05,
|
| 1254 |
+
"loss": 0.8569,
|
| 1255 |
+
"step": 16000
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 14.780812485655268,
|
| 1259 |
+
"grad_norm": 2.204672336578369,
|
| 1260 |
+
"learning_rate": 3e-05,
|
| 1261 |
+
"loss": 0.8629,
|
| 1262 |
+
"step": 16100
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 14.872618774386046,
|
| 1266 |
+
"grad_norm": 1.790323257446289,
|
| 1267 |
+
"learning_rate": 3e-05,
|
| 1268 |
+
"loss": 0.8542,
|
| 1269 |
+
"step": 16200
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 14.964425063116824,
|
| 1273 |
+
"grad_norm": 1.8623679876327515,
|
| 1274 |
+
"learning_rate": 3e-05,
|
| 1275 |
+
"loss": 0.8566,
|
| 1276 |
+
"step": 16300
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 14.999311452834519,
|
| 1280 |
+
"eval_accuracy": 0.7127714701601164,
|
| 1281 |
+
"eval_loss": 1.0913478136062622,
|
| 1282 |
+
"eval_runtime": 8.9695,
|
| 1283 |
+
"eval_samples_per_second": 55.745,
|
| 1284 |
+
"eval_steps_per_second": 7.024,
|
| 1285 |
+
"step": 16338
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 15.056231351847602,
|
| 1289 |
+
"grad_norm": 1.6198936700820923,
|
| 1290 |
+
"learning_rate": 3e-05,
|
| 1291 |
+
"loss": 0.8233,
|
| 1292 |
+
"step": 16400
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 15.14803764057838,
|
| 1296 |
+
"grad_norm": 2.117966413497925,
|
| 1297 |
+
"learning_rate": 3e-05,
|
| 1298 |
+
"loss": 0.8004,
|
| 1299 |
+
"step": 16500
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 15.239843929309158,
|
| 1303 |
+
"grad_norm": 1.9046192169189453,
|
| 1304 |
+
"learning_rate": 3e-05,
|
| 1305 |
+
"loss": 0.783,
|
| 1306 |
+
"step": 16600
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 15.331650218039936,
|
| 1310 |
+
"grad_norm": 1.8354123830795288,
|
| 1311 |
+
"learning_rate": 3e-05,
|
| 1312 |
+
"loss": 0.8123,
|
| 1313 |
+
"step": 16700
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 15.423456506770714,
|
| 1317 |
+
"grad_norm": 1.8810902833938599,
|
| 1318 |
+
"learning_rate": 3e-05,
|
| 1319 |
+
"loss": 0.8062,
|
| 1320 |
+
"step": 16800
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 15.515262795501492,
|
| 1324 |
+
"grad_norm": 2.2442831993103027,
|
| 1325 |
+
"learning_rate": 3e-05,
|
| 1326 |
+
"loss": 0.8121,
|
| 1327 |
+
"step": 16900
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 15.60706908423227,
|
| 1331 |
+
"grad_norm": 2.308647394180298,
|
| 1332 |
+
"learning_rate": 3e-05,
|
| 1333 |
+
"loss": 0.8155,
|
| 1334 |
+
"step": 17000
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 15.698875372963048,
|
| 1338 |
+
"grad_norm": 2.2714340686798096,
|
| 1339 |
+
"learning_rate": 3e-05,
|
| 1340 |
+
"loss": 0.8211,
|
| 1341 |
+
"step": 17100
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 15.790681661693826,
|
| 1345 |
+
"grad_norm": 1.9850467443466187,
|
| 1346 |
+
"learning_rate": 3e-05,
|
| 1347 |
+
"loss": 0.8054,
|
| 1348 |
+
"step": 17200
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 15.882487950424604,
|
| 1352 |
+
"grad_norm": 2.5280234813690186,
|
| 1353 |
+
"learning_rate": 3e-05,
|
| 1354 |
+
"loss": 0.8198,
|
| 1355 |
+
"step": 17300
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 15.974294239155382,
|
| 1359 |
+
"grad_norm": 2.184380292892456,
|
| 1360 |
+
"learning_rate": 3e-05,
|
| 1361 |
+
"loss": 0.8307,
|
| 1362 |
+
"step": 17400
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 16.0,
|
| 1366 |
+
"eval_accuracy": 0.7155633187772926,
|
| 1367 |
+
"eval_loss": 1.0599175691604614,
|
| 1368 |
+
"eval_runtime": 8.9521,
|
| 1369 |
+
"eval_samples_per_second": 55.853,
|
| 1370 |
+
"eval_steps_per_second": 7.037,
|
| 1371 |
+
"step": 17428
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 16.06610052788616,
|
| 1375 |
+
"grad_norm": 2.437701463699341,
|
| 1376 |
+
"learning_rate": 3e-05,
|
| 1377 |
+
"loss": 0.7609,
|
| 1378 |
+
"step": 17500
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 16.15790681661694,
|
| 1382 |
+
"grad_norm": 2.636090040206909,
|
| 1383 |
+
"learning_rate": 3e-05,
|
| 1384 |
+
"loss": 0.758,
|
| 1385 |
+
"step": 17600
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 16.249713105347716,
|
| 1389 |
+
"grad_norm": 2.1846566200256348,
|
| 1390 |
+
"learning_rate": 3e-05,
|
| 1391 |
+
"loss": 0.7439,
|
| 1392 |
+
"step": 17700
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 16.341519394078496,
|
| 1396 |
+
"grad_norm": 2.1148085594177246,
|
| 1397 |
+
"learning_rate": 3e-05,
|
| 1398 |
+
"loss": 0.7568,
|
| 1399 |
+
"step": 17800
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 16.433325682809272,
|
| 1403 |
+
"grad_norm": 1.8323599100112915,
|
| 1404 |
+
"learning_rate": 3e-05,
|
| 1405 |
+
"loss": 0.7678,
|
| 1406 |
+
"step": 17900
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 16.525131971540052,
|
| 1410 |
+
"grad_norm": 2.67404842376709,
|
| 1411 |
+
"learning_rate": 3e-05,
|
| 1412 |
+
"loss": 0.7719,
|
| 1413 |
+
"step": 18000
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 16.616938260270828,
|
| 1417 |
+
"grad_norm": 2.3159210681915283,
|
| 1418 |
+
"learning_rate": 3e-05,
|
| 1419 |
+
"loss": 0.78,
|
| 1420 |
+
"step": 18100
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 16.708744549001608,
|
| 1424 |
+
"grad_norm": 1.924141526222229,
|
| 1425 |
+
"learning_rate": 3e-05,
|
| 1426 |
+
"loss": 0.7774,
|
| 1427 |
+
"step": 18200
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 16.800550837732384,
|
| 1431 |
+
"grad_norm": 1.9718719720840454,
|
| 1432 |
+
"learning_rate": 3e-05,
|
| 1433 |
+
"loss": 0.7714,
|
| 1434 |
+
"step": 18300
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 16.892357126463164,
|
| 1438 |
+
"grad_norm": 2.0986855030059814,
|
| 1439 |
+
"learning_rate": 3e-05,
|
| 1440 |
+
"loss": 0.7861,
|
| 1441 |
+
"step": 18400
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 16.98416341519394,
|
| 1445 |
+
"grad_norm": 2.2935447692871094,
|
| 1446 |
+
"learning_rate": 3e-05,
|
| 1447 |
+
"loss": 0.7803,
|
| 1448 |
+
"step": 18500
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 16.999770484278173,
|
| 1452 |
+
"eval_accuracy": 0.7184133915574964,
|
| 1453 |
+
"eval_loss": 1.028311014175415,
|
| 1454 |
+
"eval_runtime": 8.9448,
|
| 1455 |
+
"eval_samples_per_second": 55.898,
|
| 1456 |
+
"eval_steps_per_second": 7.043,
|
| 1457 |
+
"step": 18517
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 17.07596970392472,
|
| 1461 |
+
"grad_norm": 2.364075183868408,
|
| 1462 |
+
"learning_rate": 3e-05,
|
| 1463 |
+
"loss": 0.7264,
|
| 1464 |
+
"step": 18600
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 17.167775992655496,
|
| 1468 |
+
"grad_norm": 2.1636979579925537,
|
| 1469 |
+
"learning_rate": 3e-05,
|
| 1470 |
+
"loss": 0.7038,
|
| 1471 |
+
"step": 18700
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 17.259582281386276,
|
| 1475 |
+
"grad_norm": 2.135673761367798,
|
| 1476 |
+
"learning_rate": 3e-05,
|
| 1477 |
+
"loss": 0.7136,
|
| 1478 |
+
"step": 18800
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 17.351388570117052,
|
| 1482 |
+
"grad_norm": 2.1516411304473877,
|
| 1483 |
+
"learning_rate": 3e-05,
|
| 1484 |
+
"loss": 0.7231,
|
| 1485 |
+
"step": 18900
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 17.443194858847832,
|
| 1489 |
+
"grad_norm": 2.499406337738037,
|
| 1490 |
+
"learning_rate": 3e-05,
|
| 1491 |
+
"loss": 0.7302,
|
| 1492 |
+
"step": 19000
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 17.53500114757861,
|
| 1496 |
+
"grad_norm": 2.455547332763672,
|
| 1497 |
+
"learning_rate": 3e-05,
|
| 1498 |
+
"loss": 0.7407,
|
| 1499 |
+
"step": 19100
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 17.626807436309388,
|
| 1503 |
+
"grad_norm": 2.248194932937622,
|
| 1504 |
+
"learning_rate": 3e-05,
|
| 1505 |
+
"loss": 0.722,
|
| 1506 |
+
"step": 19200
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 17.718613725040164,
|
| 1510 |
+
"grad_norm": 2.3520660400390625,
|
| 1511 |
+
"learning_rate": 3e-05,
|
| 1512 |
+
"loss": 0.7291,
|
| 1513 |
+
"step": 19300
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 17.810420013770944,
|
| 1517 |
+
"grad_norm": 2.1547889709472656,
|
| 1518 |
+
"learning_rate": 3e-05,
|
| 1519 |
+
"loss": 0.7317,
|
| 1520 |
+
"step": 19400
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 17.90222630250172,
|
| 1524 |
+
"grad_norm": 2.608548402786255,
|
| 1525 |
+
"learning_rate": 3e-05,
|
| 1526 |
+
"loss": 0.738,
|
| 1527 |
+
"step": 19500
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 17.9940325912325,
|
| 1531 |
+
"grad_norm": 2.2248220443725586,
|
| 1532 |
+
"learning_rate": 3e-05,
|
| 1533 |
+
"loss": 0.7486,
|
| 1534 |
+
"step": 19600
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 17.999540968556346,
|
| 1538 |
+
"eval_accuracy": 0.7214614264919942,
|
| 1539 |
+
"eval_loss": 0.9996564984321594,
|
| 1540 |
+
"eval_runtime": 8.9645,
|
| 1541 |
+
"eval_samples_per_second": 55.776,
|
| 1542 |
+
"eval_steps_per_second": 7.028,
|
| 1543 |
+
"step": 19606
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 18.085838879963276,
|
| 1547 |
+
"grad_norm": 2.197584867477417,
|
| 1548 |
+
"learning_rate": 3e-05,
|
| 1549 |
+
"loss": 0.6731,
|
| 1550 |
+
"step": 19700
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 18.177645168694056,
|
| 1554 |
+
"grad_norm": 2.392916440963745,
|
| 1555 |
+
"learning_rate": 3e-05,
|
| 1556 |
+
"loss": 0.676,
|
| 1557 |
+
"step": 19800
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 18.269451457424832,
|
| 1561 |
+
"grad_norm": 2.4115874767303467,
|
| 1562 |
+
"learning_rate": 3e-05,
|
| 1563 |
+
"loss": 0.691,
|
| 1564 |
+
"step": 19900
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 18.361257746155612,
|
| 1568 |
+
"grad_norm": 2.320349931716919,
|
| 1569 |
+
"learning_rate": 3e-05,
|
| 1570 |
+
"loss": 0.677,
|
| 1571 |
+
"step": 20000
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 18.45306403488639,
|
| 1575 |
+
"grad_norm": 2.2987887859344482,
|
| 1576 |
+
"learning_rate": 3e-05,
|
| 1577 |
+
"loss": 0.6857,
|
| 1578 |
+
"step": 20100
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 18.544870323617168,
|
| 1582 |
+
"grad_norm": 2.541984796524048,
|
| 1583 |
+
"learning_rate": 3e-05,
|
| 1584 |
+
"loss": 0.6787,
|
| 1585 |
+
"step": 20200
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 18.636676612347944,
|
| 1589 |
+
"grad_norm": 2.0782082080841064,
|
| 1590 |
+
"learning_rate": 3e-05,
|
| 1591 |
+
"loss": 0.6973,
|
| 1592 |
+
"step": 20300
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 18.728482901078724,
|
| 1596 |
+
"grad_norm": 2.4935009479522705,
|
| 1597 |
+
"learning_rate": 3e-05,
|
| 1598 |
+
"loss": 0.7083,
|
| 1599 |
+
"step": 20400
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 18.8202891898095,
|
| 1603 |
+
"grad_norm": 2.8205904960632324,
|
| 1604 |
+
"learning_rate": 3e-05,
|
| 1605 |
+
"loss": 0.6872,
|
| 1606 |
+
"step": 20500
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 18.91209547854028,
|
| 1610 |
+
"grad_norm": 2.335952043533325,
|
| 1611 |
+
"learning_rate": 3e-05,
|
| 1612 |
+
"loss": 0.6992,
|
| 1613 |
+
"step": 20600
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 18.99931145283452,
|
| 1617 |
+
"eval_accuracy": 0.7237583697234352,
|
| 1618 |
+
"eval_loss": 0.971889078617096,
|
| 1619 |
+
"eval_runtime": 9.1348,
|
| 1620 |
+
"eval_samples_per_second": 54.736,
|
| 1621 |
+
"eval_steps_per_second": 6.897,
|
| 1622 |
+
"step": 20695
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 19.003901767271056,
|
| 1626 |
+
"grad_norm": 1.9122214317321777,
|
| 1627 |
+
"learning_rate": 3e-05,
|
| 1628 |
+
"loss": 0.7018,
|
| 1629 |
+
"step": 20700
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 19.095708056001836,
|
| 1633 |
+
"grad_norm": 2.1178972721099854,
|
| 1634 |
+
"learning_rate": 3e-05,
|
| 1635 |
+
"loss": 0.6337,
|
| 1636 |
+
"step": 20800
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 19.187514344732612,
|
| 1640 |
+
"grad_norm": 2.1954286098480225,
|
| 1641 |
+
"learning_rate": 3e-05,
|
| 1642 |
+
"loss": 0.6294,
|
| 1643 |
+
"step": 20900
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 19.279320633463392,
|
| 1647 |
+
"grad_norm": 2.2881522178649902,
|
| 1648 |
+
"learning_rate": 3e-05,
|
| 1649 |
+
"loss": 0.6436,
|
| 1650 |
+
"step": 21000
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 19.371126922194172,
|
| 1654 |
+
"grad_norm": 2.2738537788391113,
|
| 1655 |
+
"learning_rate": 3e-05,
|
| 1656 |
+
"loss": 0.6584,
|
| 1657 |
+
"step": 21100
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 19.462933210924948,
|
| 1661 |
+
"grad_norm": 2.3467330932617188,
|
| 1662 |
+
"learning_rate": 3e-05,
|
| 1663 |
+
"loss": 0.655,
|
| 1664 |
+
"step": 21200
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 19.554739499655728,
|
| 1668 |
+
"grad_norm": 2.7984132766723633,
|
| 1669 |
+
"learning_rate": 3e-05,
|
| 1670 |
+
"loss": 0.647,
|
| 1671 |
+
"step": 21300
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"epoch": 19.646545788386504,
|
| 1675 |
+
"grad_norm": 2.397935152053833,
|
| 1676 |
+
"learning_rate": 3e-05,
|
| 1677 |
+
"loss": 0.6642,
|
| 1678 |
+
"step": 21400
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"epoch": 19.738352077117284,
|
| 1682 |
+
"grad_norm": 2.7952253818511963,
|
| 1683 |
+
"learning_rate": 3e-05,
|
| 1684 |
+
"loss": 0.6604,
|
| 1685 |
+
"step": 21500
|
| 1686 |
+
},
|
| 1687 |
+
{
|
| 1688 |
+
"epoch": 19.83015836584806,
|
| 1689 |
+
"grad_norm": 2.212345600128174,
|
| 1690 |
+
"learning_rate": 3e-05,
|
| 1691 |
+
"loss": 0.6598,
|
| 1692 |
+
"step": 21600
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 19.92196465457884,
|
| 1696 |
+
"grad_norm": 2.5237057209014893,
|
| 1697 |
+
"learning_rate": 3e-05,
|
| 1698 |
+
"loss": 0.6632,
|
| 1699 |
+
"step": 21700
|
| 1700 |
+
},
|
| 1701 |
+
{
|
| 1702 |
+
"epoch": 19.995409685563462,
|
| 1703 |
+
"eval_accuracy": 0.7263318777292577,
|
| 1704 |
+
"eval_loss": 0.9440018534660339,
|
| 1705 |
+
"eval_runtime": 8.9833,
|
| 1706 |
+
"eval_samples_per_second": 55.659,
|
| 1707 |
+
"eval_steps_per_second": 7.013,
|
| 1708 |
+
"step": 21780
|
| 1709 |
+
},
|
| 1710 |
+
{
|
| 1711 |
+
"epoch": 19.995409685563462,
|
| 1712 |
+
"step": 21780,
|
| 1713 |
+
"total_flos": 2.2953223726028554e+18,
|
| 1714 |
+
"train_loss": 1.0977098983400344,
|
| 1715 |
+
"train_runtime": 46950.9086,
|
| 1716 |
+
"train_samples_per_second": 14.847,
|
| 1717 |
+
"train_steps_per_second": 0.464
|
| 1718 |
+
}
|
| 1719 |
+
],
|
| 1720 |
+
"logging_steps": 100,
|
| 1721 |
+
"max_steps": 21780,
|
| 1722 |
+
"num_input_tokens_seen": 0,
|
| 1723 |
+
"num_train_epochs": 20,
|
| 1724 |
+
"save_steps": 500,
|
| 1725 |
+
"total_flos": 2.2953223726028554e+18,
|
| 1726 |
+
"train_batch_size": 1,
|
| 1727 |
+
"trial_name": null,
|
| 1728 |
+
"trial_params": null
|
| 1729 |
+
}
|