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 +1379 -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_docidx_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_docidx_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: 1.3258
|
| 22 |
- Accuracy: 0.7513
|
|
|
|
| 3 |
base_model: Qwen/Qwen1.5-4B
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- tyzhu/lmind_hotpot_train8000_eval7405_v1_docidx
|
| 8 |
metrics:
|
| 9 |
- accuracy
|
| 10 |
model-index:
|
| 11 |
- name: lmind_hotpot_train8000_eval7405_v1_docidx_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_docidx
|
| 18 |
+
type: tyzhu/lmind_hotpot_train8000_eval7405_v1_docidx
|
| 19 |
+
metrics:
|
| 20 |
+
- name: Accuracy
|
| 21 |
+
type: accuracy
|
| 22 |
+
value: 0.7512829373650108
|
| 23 |
library_name: peft
|
| 24 |
---
|
| 25 |
|
|
|
|
| 28 |
|
| 29 |
# lmind_hotpot_train8000_eval7405_v1_docidx_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_docidx dataset.
|
| 32 |
It achieves the following results on the evaluation set:
|
| 33 |
- Loss: 1.3258
|
| 34 |
- Accuracy: 0.7513
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.998510574918082,
|
| 3 |
+
"eval_accuracy": 0.7512829373650108,
|
| 4 |
+
"eval_loss": 1.3258286714553833,
|
| 5 |
+
"eval_runtime": 7.7118,
|
| 6 |
+
"eval_samples": 500,
|
| 7 |
+
"eval_samples_per_second": 64.836,
|
| 8 |
+
"eval_steps_per_second": 8.169,
|
| 9 |
+
"perplexity": 3.765304262709877,
|
| 10 |
+
"total_flos": 1.3733500524072796e+18,
|
| 11 |
+
"train_loss": 0.22626538049336412,
|
| 12 |
+
"train_runtime": 9900.7803,
|
| 13 |
+
"train_samples": 26854,
|
| 14 |
+
"train_samples_per_second": 54.246,
|
| 15 |
+
"train_steps_per_second": 1.695
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.998510574918082,
|
| 3 |
+
"eval_accuracy": 0.7512829373650108,
|
| 4 |
+
"eval_loss": 1.3258286714553833,
|
| 5 |
+
"eval_runtime": 7.7118,
|
| 6 |
+
"eval_samples": 500,
|
| 7 |
+
"eval_samples_per_second": 64.836,
|
| 8 |
+
"eval_steps_per_second": 8.169,
|
| 9 |
+
"perplexity": 3.765304262709877
|
| 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": 19.998510574918082,
|
| 3 |
+
"total_flos": 1.3733500524072796e+18,
|
| 4 |
+
"train_loss": 0.22626538049336412,
|
| 5 |
+
"train_runtime": 9900.7803,
|
| 6 |
+
"train_samples": 26854,
|
| 7 |
+
"train_samples_per_second": 54.246,
|
| 8 |
+
"train_steps_per_second": 1.695
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1379 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 19.998510574918082,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 16780,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.11915400655347036,
|
| 13 |
+
"grad_norm": 0.24907414615154266,
|
| 14 |
+
"learning_rate": 3e-05,
|
| 15 |
+
"loss": 1.7899,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.23830801310694072,
|
| 20 |
+
"grad_norm": 0.2722090780735016,
|
| 21 |
+
"learning_rate": 3e-05,
|
| 22 |
+
"loss": 1.6296,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.3574620196604111,
|
| 27 |
+
"grad_norm": 0.30233123898506165,
|
| 28 |
+
"learning_rate": 3e-05,
|
| 29 |
+
"loss": 1.6514,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.47661602621388144,
|
| 34 |
+
"grad_norm": 0.281762957572937,
|
| 35 |
+
"learning_rate": 3e-05,
|
| 36 |
+
"loss": 1.6225,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.5957700327673519,
|
| 41 |
+
"grad_norm": 0.2878308892250061,
|
| 42 |
+
"learning_rate": 3e-05,
|
| 43 |
+
"loss": 1.6158,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.7149240393208222,
|
| 48 |
+
"grad_norm": 0.26083847880363464,
|
| 49 |
+
"learning_rate": 3e-05,
|
| 50 |
+
"loss": 1.6232,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.8340780458742926,
|
| 55 |
+
"grad_norm": 0.3153495490550995,
|
| 56 |
+
"learning_rate": 3e-05,
|
| 57 |
+
"loss": 1.616,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.9532320524277629,
|
| 62 |
+
"grad_norm": 0.3171740770339966,
|
| 63 |
+
"learning_rate": 3e-05,
|
| 64 |
+
"loss": 1.6137,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.9997021149836163,
|
| 69 |
+
"eval_accuracy": 0.7176760259179266,
|
| 70 |
+
"eval_loss": 1.8263936042785645,
|
| 71 |
+
"eval_runtime": 7.3244,
|
| 72 |
+
"eval_samples_per_second": 68.265,
|
| 73 |
+
"eval_steps_per_second": 8.601,
|
| 74 |
+
"step": 839
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.0723860589812333,
|
| 78 |
+
"grad_norm": 0.30662721395492554,
|
| 79 |
+
"learning_rate": 3e-05,
|
| 80 |
+
"loss": 1.6034,
|
| 81 |
+
"step": 900
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.1915400655347037,
|
| 85 |
+
"grad_norm": 0.3360077738761902,
|
| 86 |
+
"learning_rate": 3e-05,
|
| 87 |
+
"loss": 1.5982,
|
| 88 |
+
"step": 1000
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.310694072088174,
|
| 92 |
+
"grad_norm": 0.42994430661201477,
|
| 93 |
+
"learning_rate": 3e-05,
|
| 94 |
+
"loss": 1.5979,
|
| 95 |
+
"step": 1100
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.4298480786416443,
|
| 99 |
+
"grad_norm": 0.37275680899620056,
|
| 100 |
+
"learning_rate": 3e-05,
|
| 101 |
+
"loss": 1.5859,
|
| 102 |
+
"step": 1200
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.5490020851951147,
|
| 106 |
+
"grad_norm": 0.3591364622116089,
|
| 107 |
+
"learning_rate": 3e-05,
|
| 108 |
+
"loss": 1.6032,
|
| 109 |
+
"step": 1300
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.668156091748585,
|
| 113 |
+
"grad_norm": 0.43212202191352844,
|
| 114 |
+
"learning_rate": 3e-05,
|
| 115 |
+
"loss": 1.5805,
|
| 116 |
+
"step": 1400
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.7873100983020556,
|
| 120 |
+
"grad_norm": 0.36169254779815674,
|
| 121 |
+
"learning_rate": 3e-05,
|
| 122 |
+
"loss": 1.5891,
|
| 123 |
+
"step": 1500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.9064641048555258,
|
| 127 |
+
"grad_norm": 0.37111684679985046,
|
| 128 |
+
"learning_rate": 3e-05,
|
| 129 |
+
"loss": 1.5978,
|
| 130 |
+
"step": 1600
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.9994042299672325,
|
| 134 |
+
"eval_accuracy": 0.7190323974082073,
|
| 135 |
+
"eval_loss": 1.8264212608337402,
|
| 136 |
+
"eval_runtime": 7.4268,
|
| 137 |
+
"eval_samples_per_second": 67.324,
|
| 138 |
+
"eval_steps_per_second": 8.483,
|
| 139 |
+
"step": 1678
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 2.025618111408996,
|
| 143 |
+
"grad_norm": 0.38414663076400757,
|
| 144 |
+
"learning_rate": 3e-05,
|
| 145 |
+
"loss": 1.5958,
|
| 146 |
+
"step": 1700
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.1447721179624666,
|
| 150 |
+
"grad_norm": 0.5014183521270752,
|
| 151 |
+
"learning_rate": 3e-05,
|
| 152 |
+
"loss": 1.5473,
|
| 153 |
+
"step": 1800
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.2639261245159368,
|
| 157 |
+
"grad_norm": 0.5397933721542358,
|
| 158 |
+
"learning_rate": 3e-05,
|
| 159 |
+
"loss": 1.5569,
|
| 160 |
+
"step": 1900
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.3830801310694074,
|
| 164 |
+
"grad_norm": 0.5639057159423828,
|
| 165 |
+
"learning_rate": 3e-05,
|
| 166 |
+
"loss": 1.567,
|
| 167 |
+
"step": 2000
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 2.5022341376228776,
|
| 171 |
+
"grad_norm": 0.5303636789321899,
|
| 172 |
+
"learning_rate": 3e-05,
|
| 173 |
+
"loss": 1.56,
|
| 174 |
+
"step": 2100
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 2.621388144176348,
|
| 178 |
+
"grad_norm": 0.5156840085983276,
|
| 179 |
+
"learning_rate": 3e-05,
|
| 180 |
+
"loss": 1.5562,
|
| 181 |
+
"step": 2200
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 2.7405421507298184,
|
| 185 |
+
"grad_norm": 0.5576587319374084,
|
| 186 |
+
"learning_rate": 3e-05,
|
| 187 |
+
"loss": 1.5442,
|
| 188 |
+
"step": 2300
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.8596961572832886,
|
| 192 |
+
"grad_norm": 0.5914539694786072,
|
| 193 |
+
"learning_rate": 3e-05,
|
| 194 |
+
"loss": 1.548,
|
| 195 |
+
"step": 2400
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.978850163836759,
|
| 199 |
+
"grad_norm": 0.5907685160636902,
|
| 200 |
+
"learning_rate": 3e-05,
|
| 201 |
+
"loss": 1.5493,
|
| 202 |
+
"step": 2500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 2.999106344950849,
|
| 206 |
+
"eval_accuracy": 0.7211187904967603,
|
| 207 |
+
"eval_loss": 1.7809884548187256,
|
| 208 |
+
"eval_runtime": 6.9826,
|
| 209 |
+
"eval_samples_per_second": 71.606,
|
| 210 |
+
"eval_steps_per_second": 9.022,
|
| 211 |
+
"step": 2517
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 3.0980041703902295,
|
| 215 |
+
"grad_norm": 0.597890317440033,
|
| 216 |
+
"learning_rate": 3e-05,
|
| 217 |
+
"loss": 1.5151,
|
| 218 |
+
"step": 2600
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 3.2171581769436997,
|
| 222 |
+
"grad_norm": 0.6472206115722656,
|
| 223 |
+
"learning_rate": 3e-05,
|
| 224 |
+
"loss": 1.5069,
|
| 225 |
+
"step": 2700
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 3.33631218349717,
|
| 229 |
+
"grad_norm": 0.603473424911499,
|
| 230 |
+
"learning_rate": 3e-05,
|
| 231 |
+
"loss": 1.5205,
|
| 232 |
+
"step": 2800
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 3.4554661900506405,
|
| 236 |
+
"grad_norm": 0.6695737242698669,
|
| 237 |
+
"learning_rate": 3e-05,
|
| 238 |
+
"loss": 1.505,
|
| 239 |
+
"step": 2900
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 3.5746201966041107,
|
| 243 |
+
"grad_norm": 0.6201011538505554,
|
| 244 |
+
"learning_rate": 3e-05,
|
| 245 |
+
"loss": 1.5005,
|
| 246 |
+
"step": 3000
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.6937742031575813,
|
| 250 |
+
"grad_norm": 0.6618828773498535,
|
| 251 |
+
"learning_rate": 3e-05,
|
| 252 |
+
"loss": 1.5139,
|
| 253 |
+
"step": 3100
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 3.8129282097110515,
|
| 257 |
+
"grad_norm": 0.7267889380455017,
|
| 258 |
+
"learning_rate": 3e-05,
|
| 259 |
+
"loss": 1.5124,
|
| 260 |
+
"step": 3200
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 3.932082216264522,
|
| 264 |
+
"grad_norm": 0.6930146217346191,
|
| 265 |
+
"learning_rate": 3e-05,
|
| 266 |
+
"loss": 1.5101,
|
| 267 |
+
"step": 3300
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 4.0,
|
| 271 |
+
"eval_accuracy": 0.7231447084233261,
|
| 272 |
+
"eval_loss": 1.7412924766540527,
|
| 273 |
+
"eval_runtime": 6.9917,
|
| 274 |
+
"eval_samples_per_second": 71.513,
|
| 275 |
+
"eval_steps_per_second": 9.011,
|
| 276 |
+
"step": 3357
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 4.051236222817992,
|
| 280 |
+
"grad_norm": 0.6705595850944519,
|
| 281 |
+
"learning_rate": 3e-05,
|
| 282 |
+
"loss": 1.488,
|
| 283 |
+
"step": 3400
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 4.1703902293714625,
|
| 287 |
+
"grad_norm": 0.707845151424408,
|
| 288 |
+
"learning_rate": 3e-05,
|
| 289 |
+
"loss": 1.4576,
|
| 290 |
+
"step": 3500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 4.289544235924933,
|
| 294 |
+
"grad_norm": 0.698554515838623,
|
| 295 |
+
"learning_rate": 3e-05,
|
| 296 |
+
"loss": 1.4619,
|
| 297 |
+
"step": 3600
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.408698242478403,
|
| 301 |
+
"grad_norm": 0.7763566970825195,
|
| 302 |
+
"learning_rate": 3e-05,
|
| 303 |
+
"loss": 1.454,
|
| 304 |
+
"step": 3700
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 4.5278522490318736,
|
| 308 |
+
"grad_norm": 0.7236441969871521,
|
| 309 |
+
"learning_rate": 3e-05,
|
| 310 |
+
"loss": 1.462,
|
| 311 |
+
"step": 3800
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 4.647006255585344,
|
| 315 |
+
"grad_norm": 0.7589882612228394,
|
| 316 |
+
"learning_rate": 3e-05,
|
| 317 |
+
"loss": 1.4766,
|
| 318 |
+
"step": 3900
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 4.766160262138815,
|
| 322 |
+
"grad_norm": 0.818758487701416,
|
| 323 |
+
"learning_rate": 3e-05,
|
| 324 |
+
"loss": 1.4625,
|
| 325 |
+
"step": 4000
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 4.885314268692285,
|
| 329 |
+
"grad_norm": 0.7571113109588623,
|
| 330 |
+
"learning_rate": 3e-05,
|
| 331 |
+
"loss": 1.4678,
|
| 332 |
+
"step": 4100
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 4.9997021149836165,
|
| 336 |
+
"eval_accuracy": 0.7247473002159828,
|
| 337 |
+
"eval_loss": 1.7400386333465576,
|
| 338 |
+
"eval_runtime": 7.2905,
|
| 339 |
+
"eval_samples_per_second": 68.582,
|
| 340 |
+
"eval_steps_per_second": 8.641,
|
| 341 |
+
"step": 4196
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 5.004468275245755,
|
| 345 |
+
"grad_norm": 0.711908221244812,
|
| 346 |
+
"learning_rate": 3e-05,
|
| 347 |
+
"loss": 1.4785,
|
| 348 |
+
"step": 4200
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 5.123622281799226,
|
| 352 |
+
"grad_norm": 0.7867583632469177,
|
| 353 |
+
"learning_rate": 3e-05,
|
| 354 |
+
"loss": 1.4144,
|
| 355 |
+
"step": 4300
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 5.242776288352696,
|
| 359 |
+
"grad_norm": 0.8279469013214111,
|
| 360 |
+
"learning_rate": 3e-05,
|
| 361 |
+
"loss": 1.4163,
|
| 362 |
+
"step": 4400
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 5.361930294906166,
|
| 366 |
+
"grad_norm": 0.7871400713920593,
|
| 367 |
+
"learning_rate": 3e-05,
|
| 368 |
+
"loss": 1.4124,
|
| 369 |
+
"step": 4500
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 5.481084301459637,
|
| 373 |
+
"grad_norm": 0.8378657102584839,
|
| 374 |
+
"learning_rate": 3e-05,
|
| 375 |
+
"loss": 1.4212,
|
| 376 |
+
"step": 4600
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 5.600238308013107,
|
| 380 |
+
"grad_norm": 0.8661020994186401,
|
| 381 |
+
"learning_rate": 3e-05,
|
| 382 |
+
"loss": 1.4213,
|
| 383 |
+
"step": 4700
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 5.719392314566577,
|
| 387 |
+
"grad_norm": 0.8194323778152466,
|
| 388 |
+
"learning_rate": 3e-05,
|
| 389 |
+
"loss": 1.4275,
|
| 390 |
+
"step": 4800
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 5.838546321120048,
|
| 394 |
+
"grad_norm": 0.861905038356781,
|
| 395 |
+
"learning_rate": 3e-05,
|
| 396 |
+
"loss": 1.4141,
|
| 397 |
+
"step": 4900
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 5.957700327673518,
|
| 401 |
+
"grad_norm": 0.8899133205413818,
|
| 402 |
+
"learning_rate": 3e-05,
|
| 403 |
+
"loss": 1.4236,
|
| 404 |
+
"step": 5000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 5.999404229967233,
|
| 408 |
+
"eval_accuracy": 0.726682505399568,
|
| 409 |
+
"eval_loss": 1.7018622159957886,
|
| 410 |
+
"eval_runtime": 7.202,
|
| 411 |
+
"eval_samples_per_second": 69.425,
|
| 412 |
+
"eval_steps_per_second": 8.748,
|
| 413 |
+
"step": 5035
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 6.076854334226988,
|
| 417 |
+
"grad_norm": 0.9570803642272949,
|
| 418 |
+
"learning_rate": 3e-05,
|
| 419 |
+
"loss": 1.3728,
|
| 420 |
+
"step": 5100
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 6.196008340780459,
|
| 424 |
+
"grad_norm": 0.9672366976737976,
|
| 425 |
+
"learning_rate": 3e-05,
|
| 426 |
+
"loss": 1.3649,
|
| 427 |
+
"step": 5200
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 6.31516234733393,
|
| 431 |
+
"grad_norm": 0.912813663482666,
|
| 432 |
+
"learning_rate": 3e-05,
|
| 433 |
+
"loss": 1.3586,
|
| 434 |
+
"step": 5300
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 6.434316353887399,
|
| 438 |
+
"grad_norm": 0.9537220597267151,
|
| 439 |
+
"learning_rate": 3e-05,
|
| 440 |
+
"loss": 1.3888,
|
| 441 |
+
"step": 5400
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 6.55347036044087,
|
| 445 |
+
"grad_norm": 1.0322316884994507,
|
| 446 |
+
"learning_rate": 3e-05,
|
| 447 |
+
"loss": 1.3805,
|
| 448 |
+
"step": 5500
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 6.67262436699434,
|
| 452 |
+
"grad_norm": 0.9503114223480225,
|
| 453 |
+
"learning_rate": 3e-05,
|
| 454 |
+
"loss": 1.3797,
|
| 455 |
+
"step": 5600
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 6.79177837354781,
|
| 459 |
+
"grad_norm": 0.980711817741394,
|
| 460 |
+
"learning_rate": 3e-05,
|
| 461 |
+
"loss": 1.372,
|
| 462 |
+
"step": 5700
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 6.910932380101281,
|
| 466 |
+
"grad_norm": 0.9597466588020325,
|
| 467 |
+
"learning_rate": 3e-05,
|
| 468 |
+
"loss": 1.3843,
|
| 469 |
+
"step": 5800
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 6.999106344950849,
|
| 473 |
+
"eval_accuracy": 0.7285701943844493,
|
| 474 |
+
"eval_loss": 1.67252779006958,
|
| 475 |
+
"eval_runtime": 7.0918,
|
| 476 |
+
"eval_samples_per_second": 70.504,
|
| 477 |
+
"eval_steps_per_second": 8.883,
|
| 478 |
+
"step": 5874
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 7.030086386654752,
|
| 482 |
+
"grad_norm": 1.0293294191360474,
|
| 483 |
+
"learning_rate": 3e-05,
|
| 484 |
+
"loss": 1.3604,
|
| 485 |
+
"step": 5900
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 7.149240393208221,
|
| 489 |
+
"grad_norm": 1.1493397951126099,
|
| 490 |
+
"learning_rate": 3e-05,
|
| 491 |
+
"loss": 1.321,
|
| 492 |
+
"step": 6000
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 7.268394399761692,
|
| 496 |
+
"grad_norm": 1.0697908401489258,
|
| 497 |
+
"learning_rate": 3e-05,
|
| 498 |
+
"loss": 1.3199,
|
| 499 |
+
"step": 6100
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 7.387548406315163,
|
| 503 |
+
"grad_norm": 0.9961449503898621,
|
| 504 |
+
"learning_rate": 3e-05,
|
| 505 |
+
"loss": 1.3055,
|
| 506 |
+
"step": 6200
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 7.506702412868632,
|
| 510 |
+
"grad_norm": 0.9932839274406433,
|
| 511 |
+
"learning_rate": 3e-05,
|
| 512 |
+
"loss": 1.3357,
|
| 513 |
+
"step": 6300
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 7.625856419422103,
|
| 517 |
+
"grad_norm": 1.1646201610565186,
|
| 518 |
+
"learning_rate": 3e-05,
|
| 519 |
+
"loss": 1.3237,
|
| 520 |
+
"step": 6400
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 7.745010425975574,
|
| 524 |
+
"grad_norm": 1.049346685409546,
|
| 525 |
+
"learning_rate": 3e-05,
|
| 526 |
+
"loss": 1.3364,
|
| 527 |
+
"step": 6500
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 7.864164432529043,
|
| 531 |
+
"grad_norm": 1.0449471473693848,
|
| 532 |
+
"learning_rate": 3e-05,
|
| 533 |
+
"loss": 1.3321,
|
| 534 |
+
"step": 6600
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 7.983318439082514,
|
| 538 |
+
"grad_norm": 1.0868715047836304,
|
| 539 |
+
"learning_rate": 3e-05,
|
| 540 |
+
"loss": 1.3481,
|
| 541 |
+
"step": 6700
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 8.0,
|
| 545 |
+
"eval_accuracy": 0.7304362850971923,
|
| 546 |
+
"eval_loss": 1.6380517482757568,
|
| 547 |
+
"eval_runtime": 7.0963,
|
| 548 |
+
"eval_samples_per_second": 70.459,
|
| 549 |
+
"eval_steps_per_second": 8.878,
|
| 550 |
+
"step": 6714
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 8.102472445635984,
|
| 554 |
+
"grad_norm": 1.170162558555603,
|
| 555 |
+
"learning_rate": 3e-05,
|
| 556 |
+
"loss": 1.2724,
|
| 557 |
+
"step": 6800
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 8.221626452189454,
|
| 561 |
+
"grad_norm": 1.0903751850128174,
|
| 562 |
+
"learning_rate": 3e-05,
|
| 563 |
+
"loss": 1.2611,
|
| 564 |
+
"step": 6900
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 8.340780458742925,
|
| 568 |
+
"grad_norm": 1.1591229438781738,
|
| 569 |
+
"learning_rate": 3e-05,
|
| 570 |
+
"loss": 1.2769,
|
| 571 |
+
"step": 7000
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 8.459934465296396,
|
| 575 |
+
"grad_norm": 1.0120233297348022,
|
| 576 |
+
"learning_rate": 3e-05,
|
| 577 |
+
"loss": 1.2806,
|
| 578 |
+
"step": 7100
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 8.579088471849866,
|
| 582 |
+
"grad_norm": 1.1763852834701538,
|
| 583 |
+
"learning_rate": 3e-05,
|
| 584 |
+
"loss": 1.3051,
|
| 585 |
+
"step": 7200
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 8.698242478403337,
|
| 589 |
+
"grad_norm": 1.1366691589355469,
|
| 590 |
+
"learning_rate": 3e-05,
|
| 591 |
+
"loss": 1.283,
|
| 592 |
+
"step": 7300
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 8.817396484956806,
|
| 596 |
+
"grad_norm": 1.1808778047561646,
|
| 597 |
+
"learning_rate": 3e-05,
|
| 598 |
+
"loss": 1.2863,
|
| 599 |
+
"step": 7400
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 8.936550491510276,
|
| 603 |
+
"grad_norm": 1.2300970554351807,
|
| 604 |
+
"learning_rate": 3e-05,
|
| 605 |
+
"loss": 1.2954,
|
| 606 |
+
"step": 7500
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 8.999702114983616,
|
| 610 |
+
"eval_accuracy": 0.7324362850971923,
|
| 611 |
+
"eval_loss": 1.6103289127349854,
|
| 612 |
+
"eval_runtime": 7.1148,
|
| 613 |
+
"eval_samples_per_second": 70.276,
|
| 614 |
+
"eval_steps_per_second": 8.855,
|
| 615 |
+
"step": 7553
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 9.055704498063747,
|
| 619 |
+
"grad_norm": 1.319324016571045,
|
| 620 |
+
"learning_rate": 3e-05,
|
| 621 |
+
"loss": 1.2613,
|
| 622 |
+
"step": 7600
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 9.174858504617218,
|
| 626 |
+
"grad_norm": 1.1479202508926392,
|
| 627 |
+
"learning_rate": 3e-05,
|
| 628 |
+
"loss": 1.2172,
|
| 629 |
+
"step": 7700
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 9.294012511170688,
|
| 633 |
+
"grad_norm": 1.147423505783081,
|
| 634 |
+
"learning_rate": 3e-05,
|
| 635 |
+
"loss": 1.2342,
|
| 636 |
+
"step": 7800
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 9.413166517724159,
|
| 640 |
+
"grad_norm": 1.2309343814849854,
|
| 641 |
+
"learning_rate": 3e-05,
|
| 642 |
+
"loss": 1.2386,
|
| 643 |
+
"step": 7900
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 9.53232052427763,
|
| 647 |
+
"grad_norm": 1.121598243713379,
|
| 648 |
+
"learning_rate": 3e-05,
|
| 649 |
+
"loss": 1.2312,
|
| 650 |
+
"step": 8000
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 9.651474530831099,
|
| 654 |
+
"grad_norm": 1.3166605234146118,
|
| 655 |
+
"learning_rate": 3e-05,
|
| 656 |
+
"loss": 1.2376,
|
| 657 |
+
"step": 8100
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 9.77062853738457,
|
| 661 |
+
"grad_norm": 1.2599198818206787,
|
| 662 |
+
"learning_rate": 3e-05,
|
| 663 |
+
"loss": 1.2457,
|
| 664 |
+
"step": 8200
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 9.88978254393804,
|
| 668 |
+
"grad_norm": 1.297348141670227,
|
| 669 |
+
"learning_rate": 3e-05,
|
| 670 |
+
"loss": 1.2426,
|
| 671 |
+
"step": 8300
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 9.999404229967233,
|
| 675 |
+
"eval_accuracy": 0.7338099352051836,
|
| 676 |
+
"eval_loss": 1.5784611701965332,
|
| 677 |
+
"eval_runtime": 7.1512,
|
| 678 |
+
"eval_samples_per_second": 69.918,
|
| 679 |
+
"eval_steps_per_second": 8.81,
|
| 680 |
+
"step": 8392
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 10.00893655049151,
|
| 684 |
+
"grad_norm": 1.33698308467865,
|
| 685 |
+
"learning_rate": 3e-05,
|
| 686 |
+
"loss": 1.2404,
|
| 687 |
+
"step": 8400
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 10.128090557044981,
|
| 691 |
+
"grad_norm": 1.469753384590149,
|
| 692 |
+
"learning_rate": 3e-05,
|
| 693 |
+
"loss": 1.1701,
|
| 694 |
+
"step": 8500
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 10.247244563598452,
|
| 698 |
+
"grad_norm": 1.8849149942398071,
|
| 699 |
+
"learning_rate": 3e-05,
|
| 700 |
+
"loss": 1.1834,
|
| 701 |
+
"step": 8600
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 10.36639857015192,
|
| 705 |
+
"grad_norm": 1.3924856185913086,
|
| 706 |
+
"learning_rate": 3e-05,
|
| 707 |
+
"loss": 1.1877,
|
| 708 |
+
"step": 8700
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 10.485552576705391,
|
| 712 |
+
"grad_norm": 1.3874456882476807,
|
| 713 |
+
"learning_rate": 3e-05,
|
| 714 |
+
"loss": 1.1965,
|
| 715 |
+
"step": 8800
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 10.604706583258862,
|
| 719 |
+
"grad_norm": 1.5605155229568481,
|
| 720 |
+
"learning_rate": 3e-05,
|
| 721 |
+
"loss": 1.1833,
|
| 722 |
+
"step": 8900
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 10.723860589812332,
|
| 726 |
+
"grad_norm": 1.2716716527938843,
|
| 727 |
+
"learning_rate": 3e-05,
|
| 728 |
+
"loss": 1.1923,
|
| 729 |
+
"step": 9000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 10.843014596365803,
|
| 733 |
+
"grad_norm": 1.434921383857727,
|
| 734 |
+
"learning_rate": 3e-05,
|
| 735 |
+
"loss": 1.2105,
|
| 736 |
+
"step": 9100
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 10.962168602919274,
|
| 740 |
+
"grad_norm": 1.3238595724105835,
|
| 741 |
+
"learning_rate": 3e-05,
|
| 742 |
+
"loss": 1.2169,
|
| 743 |
+
"step": 9200
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 10.99910634495085,
|
| 747 |
+
"eval_accuracy": 0.7355291576673866,
|
| 748 |
+
"eval_loss": 1.5435452461242676,
|
| 749 |
+
"eval_runtime": 7.5254,
|
| 750 |
+
"eval_samples_per_second": 66.442,
|
| 751 |
+
"eval_steps_per_second": 8.372,
|
| 752 |
+
"step": 9231
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 11.081322609472744,
|
| 756 |
+
"grad_norm": 1.414612054824829,
|
| 757 |
+
"learning_rate": 3e-05,
|
| 758 |
+
"loss": 1.1529,
|
| 759 |
+
"step": 9300
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 11.200476616026213,
|
| 763 |
+
"grad_norm": 1.3387949466705322,
|
| 764 |
+
"learning_rate": 3e-05,
|
| 765 |
+
"loss": 1.1356,
|
| 766 |
+
"step": 9400
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 11.319630622579684,
|
| 770 |
+
"grad_norm": 1.312972068786621,
|
| 771 |
+
"learning_rate": 3e-05,
|
| 772 |
+
"loss": 1.1461,
|
| 773 |
+
"step": 9500
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 11.438784629133155,
|
| 777 |
+
"grad_norm": 1.5637264251708984,
|
| 778 |
+
"learning_rate": 3e-05,
|
| 779 |
+
"loss": 1.1396,
|
| 780 |
+
"step": 9600
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 11.557938635686625,
|
| 784 |
+
"grad_norm": 1.419885277748108,
|
| 785 |
+
"learning_rate": 3e-05,
|
| 786 |
+
"loss": 1.16,
|
| 787 |
+
"step": 9700
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 11.677092642240096,
|
| 791 |
+
"grad_norm": 1.555853009223938,
|
| 792 |
+
"learning_rate": 3e-05,
|
| 793 |
+
"loss": 1.1438,
|
| 794 |
+
"step": 9800
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 11.796246648793566,
|
| 798 |
+
"grad_norm": 1.543087124824524,
|
| 799 |
+
"learning_rate": 3e-05,
|
| 800 |
+
"loss": 1.1596,
|
| 801 |
+
"step": 9900
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 11.915400655347035,
|
| 805 |
+
"grad_norm": 1.6761687994003296,
|
| 806 |
+
"learning_rate": 3e-05,
|
| 807 |
+
"loss": 1.167,
|
| 808 |
+
"step": 10000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 12.0,
|
| 812 |
+
"eval_accuracy": 0.7374902807775378,
|
| 813 |
+
"eval_loss": 1.5215668678283691,
|
| 814 |
+
"eval_runtime": 7.0414,
|
| 815 |
+
"eval_samples_per_second": 71.008,
|
| 816 |
+
"eval_steps_per_second": 8.947,
|
| 817 |
+
"step": 10071
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 12.034554661900506,
|
| 821 |
+
"grad_norm": 1.6120909452438354,
|
| 822 |
+
"learning_rate": 3e-05,
|
| 823 |
+
"loss": 1.1362,
|
| 824 |
+
"step": 10100
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 12.153708668453977,
|
| 828 |
+
"grad_norm": 1.432334542274475,
|
| 829 |
+
"learning_rate": 3e-05,
|
| 830 |
+
"loss": 1.0834,
|
| 831 |
+
"step": 10200
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 12.272862675007447,
|
| 835 |
+
"grad_norm": 1.400549292564392,
|
| 836 |
+
"learning_rate": 3e-05,
|
| 837 |
+
"loss": 1.096,
|
| 838 |
+
"step": 10300
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 12.392016681560918,
|
| 842 |
+
"grad_norm": 1.365933895111084,
|
| 843 |
+
"learning_rate": 3e-05,
|
| 844 |
+
"loss": 1.0918,
|
| 845 |
+
"step": 10400
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 12.511170688114388,
|
| 849 |
+
"grad_norm": 1.9454312324523926,
|
| 850 |
+
"learning_rate": 3e-05,
|
| 851 |
+
"loss": 1.0951,
|
| 852 |
+
"step": 10500
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 12.63032469466786,
|
| 856 |
+
"grad_norm": 1.6147247552871704,
|
| 857 |
+
"learning_rate": 3e-05,
|
| 858 |
+
"loss": 1.115,
|
| 859 |
+
"step": 10600
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 12.749478701221328,
|
| 863 |
+
"grad_norm": 1.5626009702682495,
|
| 864 |
+
"learning_rate": 3e-05,
|
| 865 |
+
"loss": 1.118,
|
| 866 |
+
"step": 10700
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 12.868632707774799,
|
| 870 |
+
"grad_norm": 1.6438124179840088,
|
| 871 |
+
"learning_rate": 3e-05,
|
| 872 |
+
"loss": 1.1273,
|
| 873 |
+
"step": 10800
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 12.98778671432827,
|
| 877 |
+
"grad_norm": 1.5316485166549683,
|
| 878 |
+
"learning_rate": 3e-05,
|
| 879 |
+
"loss": 1.1276,
|
| 880 |
+
"step": 10900
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 12.999702114983616,
|
| 884 |
+
"eval_accuracy": 0.7392181425485961,
|
| 885 |
+
"eval_loss": 1.4949097633361816,
|
| 886 |
+
"eval_runtime": 7.1295,
|
| 887 |
+
"eval_samples_per_second": 70.131,
|
| 888 |
+
"eval_steps_per_second": 8.837,
|
| 889 |
+
"step": 10910
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 13.10694072088174,
|
| 893 |
+
"grad_norm": 1.6530476808547974,
|
| 894 |
+
"learning_rate": 3e-05,
|
| 895 |
+
"loss": 1.0685,
|
| 896 |
+
"step": 11000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 13.22609472743521,
|
| 900 |
+
"grad_norm": 1.571428894996643,
|
| 901 |
+
"learning_rate": 3e-05,
|
| 902 |
+
"loss": 1.0333,
|
| 903 |
+
"step": 11100
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 13.345248733988681,
|
| 907 |
+
"grad_norm": 1.5977364778518677,
|
| 908 |
+
"learning_rate": 3e-05,
|
| 909 |
+
"loss": 1.0519,
|
| 910 |
+
"step": 11200
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 13.46440274054215,
|
| 914 |
+
"grad_norm": 1.8079047203063965,
|
| 915 |
+
"learning_rate": 3e-05,
|
| 916 |
+
"loss": 1.0746,
|
| 917 |
+
"step": 11300
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 13.58355674709562,
|
| 921 |
+
"grad_norm": 1.59292733669281,
|
| 922 |
+
"learning_rate": 3e-05,
|
| 923 |
+
"loss": 1.0751,
|
| 924 |
+
"step": 11400
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 13.702710753649091,
|
| 928 |
+
"grad_norm": 1.6971626281738281,
|
| 929 |
+
"learning_rate": 3e-05,
|
| 930 |
+
"loss": 1.0613,
|
| 931 |
+
"step": 11500
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 13.821864760202562,
|
| 935 |
+
"grad_norm": 1.7633167505264282,
|
| 936 |
+
"learning_rate": 3e-05,
|
| 937 |
+
"loss": 1.0639,
|
| 938 |
+
"step": 11600
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 13.941018766756033,
|
| 942 |
+
"grad_norm": 1.5861283540725708,
|
| 943 |
+
"learning_rate": 3e-05,
|
| 944 |
+
"loss": 1.0819,
|
| 945 |
+
"step": 11700
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 13.999404229967233,
|
| 949 |
+
"eval_accuracy": 0.7406004319654428,
|
| 950 |
+
"eval_loss": 1.4818942546844482,
|
| 951 |
+
"eval_runtime": 7.4071,
|
| 952 |
+
"eval_samples_per_second": 67.503,
|
| 953 |
+
"eval_steps_per_second": 8.505,
|
| 954 |
+
"step": 11749
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 14.060172773309503,
|
| 958 |
+
"grad_norm": 1.7363096475601196,
|
| 959 |
+
"learning_rate": 3e-05,
|
| 960 |
+
"loss": 1.059,
|
| 961 |
+
"step": 11800
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 14.179326779862972,
|
| 965 |
+
"grad_norm": 1.8236732482910156,
|
| 966 |
+
"learning_rate": 3e-05,
|
| 967 |
+
"loss": 1.0138,
|
| 968 |
+
"step": 11900
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 14.298480786416443,
|
| 972 |
+
"grad_norm": 1.7334719896316528,
|
| 973 |
+
"learning_rate": 3e-05,
|
| 974 |
+
"loss": 1.0069,
|
| 975 |
+
"step": 12000
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 14.417634792969913,
|
| 979 |
+
"grad_norm": 1.6551170349121094,
|
| 980 |
+
"learning_rate": 3e-05,
|
| 981 |
+
"loss": 1.0113,
|
| 982 |
+
"step": 12100
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 14.536788799523384,
|
| 986 |
+
"grad_norm": 1.630021095275879,
|
| 987 |
+
"learning_rate": 3e-05,
|
| 988 |
+
"loss": 1.0346,
|
| 989 |
+
"step": 12200
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 14.655942806076855,
|
| 993 |
+
"grad_norm": 1.7144087553024292,
|
| 994 |
+
"learning_rate": 3e-05,
|
| 995 |
+
"loss": 1.0095,
|
| 996 |
+
"step": 12300
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 14.775096812630325,
|
| 1000 |
+
"grad_norm": 1.7622939348220825,
|
| 1001 |
+
"learning_rate": 3e-05,
|
| 1002 |
+
"loss": 1.0338,
|
| 1003 |
+
"step": 12400
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 14.894250819183796,
|
| 1007 |
+
"grad_norm": 1.7320427894592285,
|
| 1008 |
+
"learning_rate": 3e-05,
|
| 1009 |
+
"loss": 1.032,
|
| 1010 |
+
"step": 12500
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 14.99910634495085,
|
| 1014 |
+
"eval_accuracy": 0.7426090712742981,
|
| 1015 |
+
"eval_loss": 1.4468342065811157,
|
| 1016 |
+
"eval_runtime": 7.1972,
|
| 1017 |
+
"eval_samples_per_second": 69.471,
|
| 1018 |
+
"eval_steps_per_second": 8.753,
|
| 1019 |
+
"step": 12588
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 15.01787310098302,
|
| 1023 |
+
"grad_norm": 1.8335142135620117,
|
| 1024 |
+
"learning_rate": 3e-05,
|
| 1025 |
+
"loss": 0.9336,
|
| 1026 |
+
"step": 12600
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 15.137027107536491,
|
| 1030 |
+
"grad_norm": 1.7047183513641357,
|
| 1031 |
+
"learning_rate": 3e-05,
|
| 1032 |
+
"loss": 0.9394,
|
| 1033 |
+
"step": 12700
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 15.256181114089962,
|
| 1037 |
+
"grad_norm": 1.5435420274734497,
|
| 1038 |
+
"learning_rate": 3e-05,
|
| 1039 |
+
"loss": 0.9804,
|
| 1040 |
+
"step": 12800
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 15.375335120643431,
|
| 1044 |
+
"grad_norm": 1.994692325592041,
|
| 1045 |
+
"learning_rate": 3e-05,
|
| 1046 |
+
"loss": 0.9765,
|
| 1047 |
+
"step": 12900
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 15.494489127196902,
|
| 1051 |
+
"grad_norm": 2.072622537612915,
|
| 1052 |
+
"learning_rate": 3e-05,
|
| 1053 |
+
"loss": 0.9955,
|
| 1054 |
+
"step": 13000
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 15.613643133750372,
|
| 1058 |
+
"grad_norm": 2.14208722114563,
|
| 1059 |
+
"learning_rate": 3e-05,
|
| 1060 |
+
"loss": 0.9814,
|
| 1061 |
+
"step": 13100
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 15.732797140303843,
|
| 1065 |
+
"grad_norm": 1.855945110321045,
|
| 1066 |
+
"learning_rate": 3e-05,
|
| 1067 |
+
"loss": 0.9939,
|
| 1068 |
+
"step": 13200
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 15.851951146857314,
|
| 1072 |
+
"grad_norm": 1.8408160209655762,
|
| 1073 |
+
"learning_rate": 3e-05,
|
| 1074 |
+
"loss": 0.9853,
|
| 1075 |
+
"step": 13300
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 15.971105153410784,
|
| 1079 |
+
"grad_norm": 1.8263169527053833,
|
| 1080 |
+
"learning_rate": 3e-05,
|
| 1081 |
+
"loss": 0.9981,
|
| 1082 |
+
"step": 13400
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 15.999702114983616,
|
| 1086 |
+
"eval_accuracy": 0.7442721382289417,
|
| 1087 |
+
"eval_loss": 1.4091745615005493,
|
| 1088 |
+
"eval_runtime": 8.0125,
|
| 1089 |
+
"eval_samples_per_second": 62.403,
|
| 1090 |
+
"eval_steps_per_second": 7.863,
|
| 1091 |
+
"step": 13424
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 16.090259159964255,
|
| 1095 |
+
"grad_norm": 2.0035765171051025,
|
| 1096 |
+
"learning_rate": 3e-05,
|
| 1097 |
+
"loss": 0.9524,
|
| 1098 |
+
"step": 13500
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 16.209413166517724,
|
| 1102 |
+
"grad_norm": 2.207390785217285,
|
| 1103 |
+
"learning_rate": 3e-05,
|
| 1104 |
+
"loss": 0.9136,
|
| 1105 |
+
"step": 13600
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 16.328567173071196,
|
| 1109 |
+
"grad_norm": 1.8691837787628174,
|
| 1110 |
+
"learning_rate": 3e-05,
|
| 1111 |
+
"loss": 0.9374,
|
| 1112 |
+
"step": 13700
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 16.447721179624665,
|
| 1116 |
+
"grad_norm": 1.9585497379302979,
|
| 1117 |
+
"learning_rate": 3e-05,
|
| 1118 |
+
"loss": 0.9427,
|
| 1119 |
+
"step": 13800
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 16.566875186178134,
|
| 1123 |
+
"grad_norm": 2.1230435371398926,
|
| 1124 |
+
"learning_rate": 3e-05,
|
| 1125 |
+
"loss": 0.9485,
|
| 1126 |
+
"step": 13900
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 16.686029192731606,
|
| 1130 |
+
"grad_norm": 1.8812588453292847,
|
| 1131 |
+
"learning_rate": 3e-05,
|
| 1132 |
+
"loss": 0.9474,
|
| 1133 |
+
"step": 14000
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 16.805183199285075,
|
| 1137 |
+
"grad_norm": 2.00522518157959,
|
| 1138 |
+
"learning_rate": 3e-05,
|
| 1139 |
+
"loss": 0.9554,
|
| 1140 |
+
"step": 14100
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 16.924337205838548,
|
| 1144 |
+
"grad_norm": 2.1199073791503906,
|
| 1145 |
+
"learning_rate": 3e-05,
|
| 1146 |
+
"loss": 0.9523,
|
| 1147 |
+
"step": 14200
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 16.999404229967233,
|
| 1151 |
+
"eval_accuracy": 0.7462505399568035,
|
| 1152 |
+
"eval_loss": 1.394910454750061,
|
| 1153 |
+
"eval_runtime": 7.7306,
|
| 1154 |
+
"eval_samples_per_second": 64.678,
|
| 1155 |
+
"eval_steps_per_second": 8.149,
|
| 1156 |
+
"step": 14263
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 17.043491212392016,
|
| 1160 |
+
"grad_norm": 1.9974786043167114,
|
| 1161 |
+
"learning_rate": 3e-05,
|
| 1162 |
+
"loss": 0.9456,
|
| 1163 |
+
"step": 14300
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 17.16264521894549,
|
| 1167 |
+
"grad_norm": 1.8176357746124268,
|
| 1168 |
+
"learning_rate": 3e-05,
|
| 1169 |
+
"loss": 0.8874,
|
| 1170 |
+
"step": 14400
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 17.281799225498958,
|
| 1174 |
+
"grad_norm": 2.0692601203918457,
|
| 1175 |
+
"learning_rate": 3e-05,
|
| 1176 |
+
"loss": 0.9041,
|
| 1177 |
+
"step": 14500
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 17.400953232052427,
|
| 1181 |
+
"grad_norm": 2.0602495670318604,
|
| 1182 |
+
"learning_rate": 3e-05,
|
| 1183 |
+
"loss": 0.9052,
|
| 1184 |
+
"step": 14600
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 17.5201072386059,
|
| 1188 |
+
"grad_norm": 2.2935659885406494,
|
| 1189 |
+
"learning_rate": 3e-05,
|
| 1190 |
+
"loss": 0.9026,
|
| 1191 |
+
"step": 14700
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 17.639261245159368,
|
| 1195 |
+
"grad_norm": 2.3279073238372803,
|
| 1196 |
+
"learning_rate": 3e-05,
|
| 1197 |
+
"loss": 0.8857,
|
| 1198 |
+
"step": 14800
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 17.75841525171284,
|
| 1202 |
+
"grad_norm": 2.5026357173919678,
|
| 1203 |
+
"learning_rate": 3e-05,
|
| 1204 |
+
"loss": 0.9163,
|
| 1205 |
+
"step": 14900
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 17.87756925826631,
|
| 1209 |
+
"grad_norm": 1.9783616065979004,
|
| 1210 |
+
"learning_rate": 3e-05,
|
| 1211 |
+
"loss": 0.907,
|
| 1212 |
+
"step": 15000
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 17.996723264819778,
|
| 1216 |
+
"grad_norm": 1.7425028085708618,
|
| 1217 |
+
"learning_rate": 3e-05,
|
| 1218 |
+
"loss": 0.9281,
|
| 1219 |
+
"step": 15100
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 17.99910634495085,
|
| 1223 |
+
"eval_accuracy": 0.7477105831533477,
|
| 1224 |
+
"eval_loss": 1.3853000402450562,
|
| 1225 |
+
"eval_runtime": 7.8211,
|
| 1226 |
+
"eval_samples_per_second": 63.93,
|
| 1227 |
+
"eval_steps_per_second": 8.055,
|
| 1228 |
+
"step": 15102
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 18.11587727137325,
|
| 1232 |
+
"grad_norm": 2.080223798751831,
|
| 1233 |
+
"learning_rate": 3e-05,
|
| 1234 |
+
"loss": 0.8379,
|
| 1235 |
+
"step": 15200
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 18.23503127792672,
|
| 1239 |
+
"grad_norm": 2.135795831680298,
|
| 1240 |
+
"learning_rate": 3e-05,
|
| 1241 |
+
"loss": 0.857,
|
| 1242 |
+
"step": 15300
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 18.35418528448019,
|
| 1246 |
+
"grad_norm": 1.939634919166565,
|
| 1247 |
+
"learning_rate": 3e-05,
|
| 1248 |
+
"loss": 0.8646,
|
| 1249 |
+
"step": 15400
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 18.47333929103366,
|
| 1253 |
+
"grad_norm": 2.035285234451294,
|
| 1254 |
+
"learning_rate": 3e-05,
|
| 1255 |
+
"loss": 0.8725,
|
| 1256 |
+
"step": 15500
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 18.592493297587133,
|
| 1260 |
+
"grad_norm": 1.9423282146453857,
|
| 1261 |
+
"learning_rate": 3e-05,
|
| 1262 |
+
"loss": 0.8773,
|
| 1263 |
+
"step": 15600
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 18.7116473041406,
|
| 1267 |
+
"grad_norm": 1.8535542488098145,
|
| 1268 |
+
"learning_rate": 3e-05,
|
| 1269 |
+
"loss": 0.8783,
|
| 1270 |
+
"step": 15700
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 18.83080131069407,
|
| 1274 |
+
"grad_norm": 1.8560757637023926,
|
| 1275 |
+
"learning_rate": 3e-05,
|
| 1276 |
+
"loss": 0.8783,
|
| 1277 |
+
"step": 15800
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 18.949955317247543,
|
| 1281 |
+
"grad_norm": 2.0741496086120605,
|
| 1282 |
+
"learning_rate": 3e-05,
|
| 1283 |
+
"loss": 0.8664,
|
| 1284 |
+
"step": 15900
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 19.0,
|
| 1288 |
+
"eval_accuracy": 0.7496069114470842,
|
| 1289 |
+
"eval_loss": 1.366912603378296,
|
| 1290 |
+
"eval_runtime": 7.6548,
|
| 1291 |
+
"eval_samples_per_second": 65.318,
|
| 1292 |
+
"eval_steps_per_second": 8.23,
|
| 1293 |
+
"step": 15942
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 19.069109323801012,
|
| 1297 |
+
"grad_norm": 2.1111717224121094,
|
| 1298 |
+
"learning_rate": 3e-05,
|
| 1299 |
+
"loss": 0.862,
|
| 1300 |
+
"step": 16000
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 19.188263330354484,
|
| 1304 |
+
"grad_norm": 2.0599365234375,
|
| 1305 |
+
"learning_rate": 3e-05,
|
| 1306 |
+
"loss": 0.8021,
|
| 1307 |
+
"step": 16100
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 19.307417336907953,
|
| 1311 |
+
"grad_norm": 1.752618432044983,
|
| 1312 |
+
"learning_rate": 3e-05,
|
| 1313 |
+
"loss": 0.8294,
|
| 1314 |
+
"step": 16200
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 19.426571343461426,
|
| 1318 |
+
"grad_norm": 2.0611188411712646,
|
| 1319 |
+
"learning_rate": 3e-05,
|
| 1320 |
+
"loss": 0.8242,
|
| 1321 |
+
"step": 16300
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 19.545725350014894,
|
| 1325 |
+
"grad_norm": 2.7928466796875,
|
| 1326 |
+
"learning_rate": 3e-05,
|
| 1327 |
+
"loss": 0.8313,
|
| 1328 |
+
"step": 16400
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 19.664879356568363,
|
| 1332 |
+
"grad_norm": 2.2697372436523438,
|
| 1333 |
+
"learning_rate": 3e-05,
|
| 1334 |
+
"loss": 0.8389,
|
| 1335 |
+
"step": 16500
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 19.784033363121836,
|
| 1339 |
+
"grad_norm": 1.9462206363677979,
|
| 1340 |
+
"learning_rate": 3e-05,
|
| 1341 |
+
"loss": 0.8413,
|
| 1342 |
+
"step": 16600
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 19.903187369675305,
|
| 1346 |
+
"grad_norm": 1.8066469430923462,
|
| 1347 |
+
"learning_rate": 3e-05,
|
| 1348 |
+
"loss": 0.8537,
|
| 1349 |
+
"step": 16700
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 19.998510574918082,
|
| 1353 |
+
"eval_accuracy": 0.7512829373650108,
|
| 1354 |
+
"eval_loss": 1.3258286714553833,
|
| 1355 |
+
"eval_runtime": 7.9417,
|
| 1356 |
+
"eval_samples_per_second": 62.959,
|
| 1357 |
+
"eval_steps_per_second": 7.933,
|
| 1358 |
+
"step": 16780
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 19.998510574918082,
|
| 1362 |
+
"step": 16780,
|
| 1363 |
+
"total_flos": 1.3733500524072796e+18,
|
| 1364 |
+
"train_loss": 0.22626538049336412,
|
| 1365 |
+
"train_runtime": 9900.7803,
|
| 1366 |
+
"train_samples_per_second": 54.246,
|
| 1367 |
+
"train_steps_per_second": 1.695
|
| 1368 |
+
}
|
| 1369 |
+
],
|
| 1370 |
+
"logging_steps": 100,
|
| 1371 |
+
"max_steps": 16780,
|
| 1372 |
+
"num_input_tokens_seen": 0,
|
| 1373 |
+
"num_train_epochs": 20,
|
| 1374 |
+
"save_steps": 500,
|
| 1375 |
+
"total_flos": 1.3733500524072796e+18,
|
| 1376 |
+
"train_batch_size": 1,
|
| 1377 |
+
"trial_name": null,
|
| 1378 |
+
"trial_params": null
|
| 1379 |
+
}
|