End of training
Browse files- README.md +14 -2
- all_results.json +15 -0
- eval_results.json +10 -0
- tokenizer.json +1 -6
- train_results.json +8 -0
- trainer_state.json +1210 -0
README.md
CHANGED
|
@@ -3,11 +3,23 @@ license: llama2
|
|
| 3 |
base_model: meta-llama/Llama-2-7b-hf
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
|
|
|
|
|
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
| 8 |
model-index:
|
| 9 |
- name: lmind_hotpot_train8000_eval7405_v1_docidx_meta-llama_Llama-2-7b-hf_3e-5_lora2
|
| 10 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
---
|
| 12 |
|
| 13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -15,7 +27,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# lmind_hotpot_train8000_eval7405_v1_docidx_meta-llama_Llama-2-7b-hf_3e-5_lora2
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 1.0143
|
| 21 |
- Accuracy: 0.7798
|
|
|
|
| 3 |
base_model: meta-llama/Llama-2-7b-hf
|
| 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_meta-llama_Llama-2-7b-hf_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: 0.7797974217311234
|
| 23 |
---
|
| 24 |
|
| 25 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 27 |
|
| 28 |
# lmind_hotpot_train8000_eval7405_v1_docidx_meta-llama_Llama-2-7b-hf_3e-5_lora2
|
| 29 |
|
| 30 |
+
This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on the tyzhu/lmind_hotpot_train8000_eval7405_v1_docidx dataset.
|
| 31 |
It achieves the following results on the evaluation set:
|
| 32 |
- Loss: 1.0143
|
| 33 |
- Accuracy: 0.7798
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.99,
|
| 3 |
+
"eval_accuracy": 0.7797974217311234,
|
| 4 |
+
"eval_loss": 1.0143389701843262,
|
| 5 |
+
"eval_runtime": 8.3752,
|
| 6 |
+
"eval_samples": 500,
|
| 7 |
+
"eval_samples_per_second": 59.7,
|
| 8 |
+
"eval_steps_per_second": 7.522,
|
| 9 |
+
"perplexity": 2.757539978631812,
|
| 10 |
+
"train_loss": 0.8347494439089824,
|
| 11 |
+
"train_runtime": 13269.4265,
|
| 12 |
+
"train_samples": 26854,
|
| 13 |
+
"train_samples_per_second": 40.475,
|
| 14 |
+
"train_steps_per_second": 1.265
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.99,
|
| 3 |
+
"eval_accuracy": 0.7797974217311234,
|
| 4 |
+
"eval_loss": 1.0143389701843262,
|
| 5 |
+
"eval_runtime": 8.3752,
|
| 6 |
+
"eval_samples": 500,
|
| 7 |
+
"eval_samples_per_second": 59.7,
|
| 8 |
+
"eval_steps_per_second": 7.522,
|
| 9 |
+
"perplexity": 2.757539978631812
|
| 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 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.99,
|
| 3 |
+
"train_loss": 0.8347494439089824,
|
| 4 |
+
"train_runtime": 13269.4265,
|
| 5 |
+
"train_samples": 26854,
|
| 6 |
+
"train_samples_per_second": 40.475,
|
| 7 |
+
"train_steps_per_second": 1.265
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1210 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 19.994042299672326,
|
| 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": 0.12,
|
| 13 |
+
"learning_rate": 3e-05,
|
| 14 |
+
"loss": 1.3583,
|
| 15 |
+
"step": 100
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.24,
|
| 19 |
+
"learning_rate": 3e-05,
|
| 20 |
+
"loss": 1.1863,
|
| 21 |
+
"step": 200
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.36,
|
| 25 |
+
"learning_rate": 3e-05,
|
| 26 |
+
"loss": 1.1667,
|
| 27 |
+
"step": 300
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.48,
|
| 31 |
+
"learning_rate": 3e-05,
|
| 32 |
+
"loss": 1.1149,
|
| 33 |
+
"step": 400
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.6,
|
| 37 |
+
"learning_rate": 3e-05,
|
| 38 |
+
"loss": 1.1321,
|
| 39 |
+
"step": 500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.71,
|
| 43 |
+
"learning_rate": 3e-05,
|
| 44 |
+
"loss": 1.1209,
|
| 45 |
+
"step": 600
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.83,
|
| 49 |
+
"learning_rate": 3e-05,
|
| 50 |
+
"loss": 1.121,
|
| 51 |
+
"step": 700
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.95,
|
| 55 |
+
"learning_rate": 3e-05,
|
| 56 |
+
"loss": 1.1149,
|
| 57 |
+
"step": 800
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 1.0,
|
| 61 |
+
"eval_accuracy": 0.7521546961325967,
|
| 62 |
+
"eval_loss": 1.3353701829910278,
|
| 63 |
+
"eval_runtime": 7.454,
|
| 64 |
+
"eval_samples_per_second": 67.078,
|
| 65 |
+
"eval_steps_per_second": 8.452,
|
| 66 |
+
"step": 839
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.07,
|
| 70 |
+
"learning_rate": 3e-05,
|
| 71 |
+
"loss": 1.1126,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.19,
|
| 76 |
+
"learning_rate": 3e-05,
|
| 77 |
+
"loss": 1.1057,
|
| 78 |
+
"step": 1000
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.31,
|
| 82 |
+
"learning_rate": 3e-05,
|
| 83 |
+
"loss": 1.0988,
|
| 84 |
+
"step": 1100
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.43,
|
| 88 |
+
"learning_rate": 3e-05,
|
| 89 |
+
"loss": 1.098,
|
| 90 |
+
"step": 1200
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.55,
|
| 94 |
+
"learning_rate": 3e-05,
|
| 95 |
+
"loss": 1.0943,
|
| 96 |
+
"step": 1300
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.67,
|
| 100 |
+
"learning_rate": 3e-05,
|
| 101 |
+
"loss": 1.1121,
|
| 102 |
+
"step": 1400
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.79,
|
| 106 |
+
"learning_rate": 3e-05,
|
| 107 |
+
"loss": 1.0936,
|
| 108 |
+
"step": 1500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.91,
|
| 112 |
+
"learning_rate": 3e-05,
|
| 113 |
+
"loss": 1.0834,
|
| 114 |
+
"step": 1600
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 2.0,
|
| 118 |
+
"eval_accuracy": 0.7535874769797422,
|
| 119 |
+
"eval_loss": 1.3234003782272339,
|
| 120 |
+
"eval_runtime": 8.165,
|
| 121 |
+
"eval_samples_per_second": 61.237,
|
| 122 |
+
"eval_steps_per_second": 7.716,
|
| 123 |
+
"step": 1678
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 2.03,
|
| 127 |
+
"learning_rate": 3e-05,
|
| 128 |
+
"loss": 1.095,
|
| 129 |
+
"step": 1700
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 2.14,
|
| 133 |
+
"learning_rate": 3e-05,
|
| 134 |
+
"loss": 1.0865,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 2.26,
|
| 139 |
+
"learning_rate": 3e-05,
|
| 140 |
+
"loss": 1.0654,
|
| 141 |
+
"step": 1900
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 2.38,
|
| 145 |
+
"learning_rate": 3e-05,
|
| 146 |
+
"loss": 1.0803,
|
| 147 |
+
"step": 2000
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.5,
|
| 151 |
+
"learning_rate": 3e-05,
|
| 152 |
+
"loss": 1.0781,
|
| 153 |
+
"step": 2100
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.62,
|
| 157 |
+
"learning_rate": 3e-05,
|
| 158 |
+
"loss": 1.0676,
|
| 159 |
+
"step": 2200
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 2.74,
|
| 163 |
+
"learning_rate": 3e-05,
|
| 164 |
+
"loss": 1.0624,
|
| 165 |
+
"step": 2300
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.86,
|
| 169 |
+
"learning_rate": 3e-05,
|
| 170 |
+
"loss": 1.078,
|
| 171 |
+
"step": 2400
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 2.98,
|
| 175 |
+
"learning_rate": 3e-05,
|
| 176 |
+
"loss": 1.0714,
|
| 177 |
+
"step": 2500
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 3.0,
|
| 181 |
+
"eval_accuracy": 0.7554180478821363,
|
| 182 |
+
"eval_loss": 1.275762677192688,
|
| 183 |
+
"eval_runtime": 8.3175,
|
| 184 |
+
"eval_samples_per_second": 60.114,
|
| 185 |
+
"eval_steps_per_second": 7.574,
|
| 186 |
+
"step": 2517
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 3.1,
|
| 190 |
+
"learning_rate": 3e-05,
|
| 191 |
+
"loss": 1.0463,
|
| 192 |
+
"step": 2600
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 3.22,
|
| 196 |
+
"learning_rate": 3e-05,
|
| 197 |
+
"loss": 1.0363,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 3.34,
|
| 202 |
+
"learning_rate": 3e-05,
|
| 203 |
+
"loss": 1.0321,
|
| 204 |
+
"step": 2800
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 3.46,
|
| 208 |
+
"learning_rate": 3e-05,
|
| 209 |
+
"loss": 1.0468,
|
| 210 |
+
"step": 2900
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 3.57,
|
| 214 |
+
"learning_rate": 3e-05,
|
| 215 |
+
"loss": 1.0447,
|
| 216 |
+
"step": 3000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 3.69,
|
| 220 |
+
"learning_rate": 3e-05,
|
| 221 |
+
"loss": 1.0194,
|
| 222 |
+
"step": 3100
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.81,
|
| 226 |
+
"learning_rate": 3e-05,
|
| 227 |
+
"loss": 1.0472,
|
| 228 |
+
"step": 3200
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 3.93,
|
| 232 |
+
"learning_rate": 3e-05,
|
| 233 |
+
"loss": 1.0547,
|
| 234 |
+
"step": 3300
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 4.0,
|
| 238 |
+
"eval_accuracy": 0.7570202578268876,
|
| 239 |
+
"eval_loss": 1.280945897102356,
|
| 240 |
+
"eval_runtime": 8.1762,
|
| 241 |
+
"eval_samples_per_second": 61.153,
|
| 242 |
+
"eval_steps_per_second": 7.705,
|
| 243 |
+
"step": 3357
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 4.05,
|
| 247 |
+
"learning_rate": 3e-05,
|
| 248 |
+
"loss": 1.0259,
|
| 249 |
+
"step": 3400
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 4.17,
|
| 253 |
+
"learning_rate": 3e-05,
|
| 254 |
+
"loss": 1.0004,
|
| 255 |
+
"step": 3500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 4.29,
|
| 259 |
+
"learning_rate": 3e-05,
|
| 260 |
+
"loss": 0.9957,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 4.41,
|
| 265 |
+
"learning_rate": 3e-05,
|
| 266 |
+
"loss": 1.0203,
|
| 267 |
+
"step": 3700
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 4.53,
|
| 271 |
+
"learning_rate": 3e-05,
|
| 272 |
+
"loss": 1.015,
|
| 273 |
+
"step": 3800
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 4.65,
|
| 277 |
+
"learning_rate": 3e-05,
|
| 278 |
+
"loss": 0.9979,
|
| 279 |
+
"step": 3900
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 4.77,
|
| 283 |
+
"learning_rate": 3e-05,
|
| 284 |
+
"loss": 1.0172,
|
| 285 |
+
"step": 4000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 4.89,
|
| 289 |
+
"learning_rate": 3e-05,
|
| 290 |
+
"loss": 1.0218,
|
| 291 |
+
"step": 4100
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 5.0,
|
| 295 |
+
"eval_accuracy": 0.7587403314917127,
|
| 296 |
+
"eval_loss": 1.255967140197754,
|
| 297 |
+
"eval_runtime": 8.3434,
|
| 298 |
+
"eval_samples_per_second": 59.928,
|
| 299 |
+
"eval_steps_per_second": 7.551,
|
| 300 |
+
"step": 4196
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 5.0,
|
| 304 |
+
"learning_rate": 3e-05,
|
| 305 |
+
"loss": 1.0096,
|
| 306 |
+
"step": 4200
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 5.12,
|
| 310 |
+
"learning_rate": 3e-05,
|
| 311 |
+
"loss": 0.962,
|
| 312 |
+
"step": 4300
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 5.24,
|
| 316 |
+
"learning_rate": 3e-05,
|
| 317 |
+
"loss": 0.9764,
|
| 318 |
+
"step": 4400
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 5.36,
|
| 322 |
+
"learning_rate": 3e-05,
|
| 323 |
+
"loss": 0.9881,
|
| 324 |
+
"step": 4500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 5.48,
|
| 328 |
+
"learning_rate": 3e-05,
|
| 329 |
+
"loss": 0.989,
|
| 330 |
+
"step": 4600
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 5.6,
|
| 334 |
+
"learning_rate": 3e-05,
|
| 335 |
+
"loss": 0.9742,
|
| 336 |
+
"step": 4700
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 5.72,
|
| 340 |
+
"learning_rate": 3e-05,
|
| 341 |
+
"loss": 0.9702,
|
| 342 |
+
"step": 4800
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 5.84,
|
| 346 |
+
"learning_rate": 3e-05,
|
| 347 |
+
"loss": 0.9717,
|
| 348 |
+
"step": 4900
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 5.96,
|
| 352 |
+
"learning_rate": 3e-05,
|
| 353 |
+
"loss": 0.9748,
|
| 354 |
+
"step": 5000
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 6.0,
|
| 358 |
+
"eval_accuracy": 0.7605119705340699,
|
| 359 |
+
"eval_loss": 1.2244011163711548,
|
| 360 |
+
"eval_runtime": 8.3507,
|
| 361 |
+
"eval_samples_per_second": 59.875,
|
| 362 |
+
"eval_steps_per_second": 7.544,
|
| 363 |
+
"step": 5035
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 6.08,
|
| 367 |
+
"learning_rate": 3e-05,
|
| 368 |
+
"loss": 0.9662,
|
| 369 |
+
"step": 5100
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 6.2,
|
| 373 |
+
"learning_rate": 3e-05,
|
| 374 |
+
"loss": 0.9357,
|
| 375 |
+
"step": 5200
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 6.32,
|
| 379 |
+
"learning_rate": 3e-05,
|
| 380 |
+
"loss": 0.9343,
|
| 381 |
+
"step": 5300
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 6.43,
|
| 385 |
+
"learning_rate": 3e-05,
|
| 386 |
+
"loss": 0.963,
|
| 387 |
+
"step": 5400
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 6.55,
|
| 391 |
+
"learning_rate": 3e-05,
|
| 392 |
+
"loss": 0.9259,
|
| 393 |
+
"step": 5500
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 6.67,
|
| 397 |
+
"learning_rate": 3e-05,
|
| 398 |
+
"loss": 0.9363,
|
| 399 |
+
"step": 5600
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 6.79,
|
| 403 |
+
"learning_rate": 3e-05,
|
| 404 |
+
"loss": 0.9612,
|
| 405 |
+
"step": 5700
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 6.91,
|
| 409 |
+
"learning_rate": 3e-05,
|
| 410 |
+
"loss": 0.9439,
|
| 411 |
+
"step": 5800
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 7.0,
|
| 415 |
+
"eval_accuracy": 0.7620920810313075,
|
| 416 |
+
"eval_loss": 1.1968433856964111,
|
| 417 |
+
"eval_runtime": 8.3266,
|
| 418 |
+
"eval_samples_per_second": 60.049,
|
| 419 |
+
"eval_steps_per_second": 7.566,
|
| 420 |
+
"step": 5874
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 7.03,
|
| 424 |
+
"learning_rate": 3e-05,
|
| 425 |
+
"loss": 0.9399,
|
| 426 |
+
"step": 5900
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 7.15,
|
| 430 |
+
"learning_rate": 3e-05,
|
| 431 |
+
"loss": 0.8941,
|
| 432 |
+
"step": 6000
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 7.27,
|
| 436 |
+
"learning_rate": 3e-05,
|
| 437 |
+
"loss": 0.9119,
|
| 438 |
+
"step": 6100
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 7.39,
|
| 442 |
+
"learning_rate": 3e-05,
|
| 443 |
+
"loss": 0.904,
|
| 444 |
+
"step": 6200
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 7.51,
|
| 448 |
+
"learning_rate": 3e-05,
|
| 449 |
+
"loss": 0.918,
|
| 450 |
+
"step": 6300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 7.63,
|
| 454 |
+
"learning_rate": 3e-05,
|
| 455 |
+
"loss": 0.9058,
|
| 456 |
+
"step": 6400
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 7.75,
|
| 460 |
+
"learning_rate": 3e-05,
|
| 461 |
+
"loss": 0.9209,
|
| 462 |
+
"step": 6500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 7.86,
|
| 466 |
+
"learning_rate": 3e-05,
|
| 467 |
+
"loss": 0.9208,
|
| 468 |
+
"step": 6600
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 7.98,
|
| 472 |
+
"learning_rate": 3e-05,
|
| 473 |
+
"loss": 0.9155,
|
| 474 |
+
"step": 6700
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 8.0,
|
| 478 |
+
"eval_accuracy": 0.763414364640884,
|
| 479 |
+
"eval_loss": 1.188802719116211,
|
| 480 |
+
"eval_runtime": 8.1247,
|
| 481 |
+
"eval_samples_per_second": 61.54,
|
| 482 |
+
"eval_steps_per_second": 7.754,
|
| 483 |
+
"step": 6714
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 8.1,
|
| 487 |
+
"learning_rate": 3e-05,
|
| 488 |
+
"loss": 0.8863,
|
| 489 |
+
"step": 6800
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 8.22,
|
| 493 |
+
"learning_rate": 3e-05,
|
| 494 |
+
"loss": 0.8825,
|
| 495 |
+
"step": 6900
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 8.34,
|
| 499 |
+
"learning_rate": 3e-05,
|
| 500 |
+
"loss": 0.879,
|
| 501 |
+
"step": 7000
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 8.46,
|
| 505 |
+
"learning_rate": 3e-05,
|
| 506 |
+
"loss": 0.8803,
|
| 507 |
+
"step": 7100
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 8.58,
|
| 511 |
+
"learning_rate": 3e-05,
|
| 512 |
+
"loss": 0.8799,
|
| 513 |
+
"step": 7200
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 8.7,
|
| 517 |
+
"learning_rate": 3e-05,
|
| 518 |
+
"loss": 0.8762,
|
| 519 |
+
"step": 7300
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 8.82,
|
| 523 |
+
"learning_rate": 3e-05,
|
| 524 |
+
"loss": 0.8734,
|
| 525 |
+
"step": 7400
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 8.94,
|
| 529 |
+
"learning_rate": 3e-05,
|
| 530 |
+
"loss": 0.8767,
|
| 531 |
+
"step": 7500
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 9.0,
|
| 535 |
+
"eval_accuracy": 0.765060773480663,
|
| 536 |
+
"eval_loss": 1.162414312362671,
|
| 537 |
+
"eval_runtime": 7.396,
|
| 538 |
+
"eval_samples_per_second": 67.604,
|
| 539 |
+
"eval_steps_per_second": 8.518,
|
| 540 |
+
"step": 7553
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 9.06,
|
| 544 |
+
"learning_rate": 3e-05,
|
| 545 |
+
"loss": 0.8476,
|
| 546 |
+
"step": 7600
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 9.17,
|
| 550 |
+
"learning_rate": 3e-05,
|
| 551 |
+
"loss": 0.8406,
|
| 552 |
+
"step": 7700
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 9.29,
|
| 556 |
+
"learning_rate": 3e-05,
|
| 557 |
+
"loss": 0.8402,
|
| 558 |
+
"step": 7800
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 9.41,
|
| 562 |
+
"learning_rate": 3e-05,
|
| 563 |
+
"loss": 0.8426,
|
| 564 |
+
"step": 7900
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 9.53,
|
| 568 |
+
"learning_rate": 3e-05,
|
| 569 |
+
"loss": 0.8619,
|
| 570 |
+
"step": 8000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 9.65,
|
| 574 |
+
"learning_rate": 3e-05,
|
| 575 |
+
"loss": 0.8431,
|
| 576 |
+
"step": 8100
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 9.77,
|
| 580 |
+
"learning_rate": 3e-05,
|
| 581 |
+
"loss": 0.8529,
|
| 582 |
+
"step": 8200
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 9.89,
|
| 586 |
+
"learning_rate": 3e-05,
|
| 587 |
+
"loss": 0.848,
|
| 588 |
+
"step": 8300
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 10.0,
|
| 592 |
+
"eval_accuracy": 0.7663646408839779,
|
| 593 |
+
"eval_loss": 1.1409990787506104,
|
| 594 |
+
"eval_runtime": 7.3191,
|
| 595 |
+
"eval_samples_per_second": 68.314,
|
| 596 |
+
"eval_steps_per_second": 8.608,
|
| 597 |
+
"step": 8392
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 10.01,
|
| 601 |
+
"learning_rate": 3e-05,
|
| 602 |
+
"loss": 0.8475,
|
| 603 |
+
"step": 8400
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 10.13,
|
| 607 |
+
"learning_rate": 3e-05,
|
| 608 |
+
"loss": 0.7954,
|
| 609 |
+
"step": 8500
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 10.25,
|
| 613 |
+
"learning_rate": 3e-05,
|
| 614 |
+
"loss": 0.8015,
|
| 615 |
+
"step": 8600
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 10.37,
|
| 619 |
+
"learning_rate": 3e-05,
|
| 620 |
+
"loss": 0.8079,
|
| 621 |
+
"step": 8700
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 10.49,
|
| 625 |
+
"learning_rate": 3e-05,
|
| 626 |
+
"loss": 0.8152,
|
| 627 |
+
"step": 8800
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 10.6,
|
| 631 |
+
"learning_rate": 3e-05,
|
| 632 |
+
"loss": 0.8236,
|
| 633 |
+
"step": 8900
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 10.72,
|
| 637 |
+
"learning_rate": 3e-05,
|
| 638 |
+
"loss": 0.8189,
|
| 639 |
+
"step": 9000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 10.84,
|
| 643 |
+
"learning_rate": 3e-05,
|
| 644 |
+
"loss": 0.8097,
|
| 645 |
+
"step": 9100
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 10.96,
|
| 649 |
+
"learning_rate": 3e-05,
|
| 650 |
+
"loss": 0.8227,
|
| 651 |
+
"step": 9200
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 11.0,
|
| 655 |
+
"eval_accuracy": 0.7677826887661142,
|
| 656 |
+
"eval_loss": 1.1287797689437866,
|
| 657 |
+
"eval_runtime": 8.3999,
|
| 658 |
+
"eval_samples_per_second": 59.525,
|
| 659 |
+
"eval_steps_per_second": 7.5,
|
| 660 |
+
"step": 9231
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 11.08,
|
| 664 |
+
"learning_rate": 3e-05,
|
| 665 |
+
"loss": 0.788,
|
| 666 |
+
"step": 9300
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 11.2,
|
| 670 |
+
"learning_rate": 3e-05,
|
| 671 |
+
"loss": 0.7663,
|
| 672 |
+
"step": 9400
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 11.32,
|
| 676 |
+
"learning_rate": 3e-05,
|
| 677 |
+
"loss": 0.7754,
|
| 678 |
+
"step": 9500
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 11.44,
|
| 682 |
+
"learning_rate": 3e-05,
|
| 683 |
+
"loss": 0.781,
|
| 684 |
+
"step": 9600
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 11.56,
|
| 688 |
+
"learning_rate": 3e-05,
|
| 689 |
+
"loss": 0.7877,
|
| 690 |
+
"step": 9700
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 11.68,
|
| 694 |
+
"learning_rate": 3e-05,
|
| 695 |
+
"loss": 0.7908,
|
| 696 |
+
"step": 9800
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 11.8,
|
| 700 |
+
"learning_rate": 3e-05,
|
| 701 |
+
"loss": 0.7729,
|
| 702 |
+
"step": 9900
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 11.92,
|
| 706 |
+
"learning_rate": 3e-05,
|
| 707 |
+
"loss": 0.7874,
|
| 708 |
+
"step": 10000
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 12.0,
|
| 712 |
+
"eval_accuracy": 0.7694217311233886,
|
| 713 |
+
"eval_loss": 1.1165099143981934,
|
| 714 |
+
"eval_runtime": 8.5032,
|
| 715 |
+
"eval_samples_per_second": 58.802,
|
| 716 |
+
"eval_steps_per_second": 7.409,
|
| 717 |
+
"step": 10071
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 12.03,
|
| 721 |
+
"learning_rate": 3e-05,
|
| 722 |
+
"loss": 0.7746,
|
| 723 |
+
"step": 10100
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 12.15,
|
| 727 |
+
"learning_rate": 3e-05,
|
| 728 |
+
"loss": 0.7404,
|
| 729 |
+
"step": 10200
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 12.27,
|
| 733 |
+
"learning_rate": 3e-05,
|
| 734 |
+
"loss": 0.7445,
|
| 735 |
+
"step": 10300
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 12.39,
|
| 739 |
+
"learning_rate": 3e-05,
|
| 740 |
+
"loss": 0.7466,
|
| 741 |
+
"step": 10400
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 12.51,
|
| 745 |
+
"learning_rate": 3e-05,
|
| 746 |
+
"loss": 0.7659,
|
| 747 |
+
"step": 10500
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 12.63,
|
| 751 |
+
"learning_rate": 3e-05,
|
| 752 |
+
"loss": 0.7515,
|
| 753 |
+
"step": 10600
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 12.75,
|
| 757 |
+
"learning_rate": 3e-05,
|
| 758 |
+
"loss": 0.7566,
|
| 759 |
+
"step": 10700
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 12.87,
|
| 763 |
+
"learning_rate": 3e-05,
|
| 764 |
+
"loss": 0.737,
|
| 765 |
+
"step": 10800
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 12.99,
|
| 769 |
+
"learning_rate": 3e-05,
|
| 770 |
+
"loss": 0.7469,
|
| 771 |
+
"step": 10900
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 13.0,
|
| 775 |
+
"eval_accuracy": 0.7703278084714549,
|
| 776 |
+
"eval_loss": 1.1008473634719849,
|
| 777 |
+
"eval_runtime": 8.2857,
|
| 778 |
+
"eval_samples_per_second": 60.345,
|
| 779 |
+
"eval_steps_per_second": 7.603,
|
| 780 |
+
"step": 10910
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 13.11,
|
| 784 |
+
"learning_rate": 3e-05,
|
| 785 |
+
"loss": 0.711,
|
| 786 |
+
"step": 11000
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 13.23,
|
| 790 |
+
"learning_rate": 3e-05,
|
| 791 |
+
"loss": 0.7035,
|
| 792 |
+
"step": 11100
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 13.35,
|
| 796 |
+
"learning_rate": 3e-05,
|
| 797 |
+
"loss": 0.7052,
|
| 798 |
+
"step": 11200
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 13.46,
|
| 802 |
+
"learning_rate": 3e-05,
|
| 803 |
+
"loss": 0.704,
|
| 804 |
+
"step": 11300
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 13.58,
|
| 808 |
+
"learning_rate": 3e-05,
|
| 809 |
+
"loss": 0.7169,
|
| 810 |
+
"step": 11400
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 13.7,
|
| 814 |
+
"learning_rate": 3e-05,
|
| 815 |
+
"loss": 0.7312,
|
| 816 |
+
"step": 11500
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 13.82,
|
| 820 |
+
"learning_rate": 3e-05,
|
| 821 |
+
"loss": 0.7334,
|
| 822 |
+
"step": 11600
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 13.94,
|
| 826 |
+
"learning_rate": 3e-05,
|
| 827 |
+
"loss": 0.7256,
|
| 828 |
+
"step": 11700
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 14.0,
|
| 832 |
+
"eval_accuracy": 0.7720662983425415,
|
| 833 |
+
"eval_loss": 1.0892044305801392,
|
| 834 |
+
"eval_runtime": 8.1389,
|
| 835 |
+
"eval_samples_per_second": 61.434,
|
| 836 |
+
"eval_steps_per_second": 7.741,
|
| 837 |
+
"step": 11749
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 14.06,
|
| 841 |
+
"learning_rate": 3e-05,
|
| 842 |
+
"loss": 0.7155,
|
| 843 |
+
"step": 11800
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 14.18,
|
| 847 |
+
"learning_rate": 3e-05,
|
| 848 |
+
"loss": 0.6688,
|
| 849 |
+
"step": 11900
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 14.3,
|
| 853 |
+
"learning_rate": 3e-05,
|
| 854 |
+
"loss": 0.6699,
|
| 855 |
+
"step": 12000
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 14.42,
|
| 859 |
+
"learning_rate": 3e-05,
|
| 860 |
+
"loss": 0.6883,
|
| 861 |
+
"step": 12100
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 14.54,
|
| 865 |
+
"learning_rate": 3e-05,
|
| 866 |
+
"loss": 0.6744,
|
| 867 |
+
"step": 12200
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 14.66,
|
| 871 |
+
"learning_rate": 3e-05,
|
| 872 |
+
"loss": 0.6957,
|
| 873 |
+
"step": 12300
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 14.78,
|
| 877 |
+
"learning_rate": 3e-05,
|
| 878 |
+
"loss": 0.6868,
|
| 879 |
+
"step": 12400
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 14.89,
|
| 883 |
+
"learning_rate": 3e-05,
|
| 884 |
+
"loss": 0.701,
|
| 885 |
+
"step": 12500
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 15.0,
|
| 889 |
+
"eval_accuracy": 0.7732338858195211,
|
| 890 |
+
"eval_loss": 1.0651440620422363,
|
| 891 |
+
"eval_runtime": 7.488,
|
| 892 |
+
"eval_samples_per_second": 66.773,
|
| 893 |
+
"eval_steps_per_second": 8.413,
|
| 894 |
+
"step": 12588
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 15.01,
|
| 898 |
+
"learning_rate": 3e-05,
|
| 899 |
+
"loss": 0.7009,
|
| 900 |
+
"step": 12600
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 15.13,
|
| 904 |
+
"learning_rate": 3e-05,
|
| 905 |
+
"loss": 0.629,
|
| 906 |
+
"step": 12700
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 15.25,
|
| 910 |
+
"learning_rate": 3e-05,
|
| 911 |
+
"loss": 0.6525,
|
| 912 |
+
"step": 12800
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 15.37,
|
| 916 |
+
"learning_rate": 3e-05,
|
| 917 |
+
"loss": 0.657,
|
| 918 |
+
"step": 12900
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 15.49,
|
| 922 |
+
"learning_rate": 3e-05,
|
| 923 |
+
"loss": 0.6568,
|
| 924 |
+
"step": 13000
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 15.61,
|
| 928 |
+
"learning_rate": 3e-05,
|
| 929 |
+
"loss": 0.6768,
|
| 930 |
+
"step": 13100
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 15.73,
|
| 934 |
+
"learning_rate": 3e-05,
|
| 935 |
+
"loss": 0.6576,
|
| 936 |
+
"step": 13200
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 15.85,
|
| 940 |
+
"learning_rate": 3e-05,
|
| 941 |
+
"loss": 0.6612,
|
| 942 |
+
"step": 13300
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 15.97,
|
| 946 |
+
"learning_rate": 3e-05,
|
| 947 |
+
"loss": 0.6651,
|
| 948 |
+
"step": 13400
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 16.0,
|
| 952 |
+
"eval_accuracy": 0.7745524861878453,
|
| 953 |
+
"eval_loss": 1.0723310708999634,
|
| 954 |
+
"eval_runtime": 8.392,
|
| 955 |
+
"eval_samples_per_second": 59.58,
|
| 956 |
+
"eval_steps_per_second": 7.507,
|
| 957 |
+
"step": 13428
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 16.09,
|
| 961 |
+
"learning_rate": 3e-05,
|
| 962 |
+
"loss": 0.6276,
|
| 963 |
+
"step": 13500
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 16.2,
|
| 967 |
+
"learning_rate": 3e-05,
|
| 968 |
+
"loss": 0.6229,
|
| 969 |
+
"step": 13600
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 16.32,
|
| 973 |
+
"learning_rate": 3e-05,
|
| 974 |
+
"loss": 0.6188,
|
| 975 |
+
"step": 13700
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 16.44,
|
| 979 |
+
"learning_rate": 3e-05,
|
| 980 |
+
"loss": 0.6215,
|
| 981 |
+
"step": 13800
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 16.56,
|
| 985 |
+
"learning_rate": 3e-05,
|
| 986 |
+
"loss": 0.6283,
|
| 987 |
+
"step": 13900
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 16.68,
|
| 991 |
+
"learning_rate": 3e-05,
|
| 992 |
+
"loss": 0.6286,
|
| 993 |
+
"step": 14000
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 16.8,
|
| 997 |
+
"learning_rate": 3e-05,
|
| 998 |
+
"loss": 0.6309,
|
| 999 |
+
"step": 14100
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 16.92,
|
| 1003 |
+
"learning_rate": 3e-05,
|
| 1004 |
+
"loss": 0.6534,
|
| 1005 |
+
"step": 14200
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 17.0,
|
| 1009 |
+
"eval_accuracy": 0.7759263351749539,
|
| 1010 |
+
"eval_loss": 1.0388367176055908,
|
| 1011 |
+
"eval_runtime": 8.4402,
|
| 1012 |
+
"eval_samples_per_second": 59.24,
|
| 1013 |
+
"eval_steps_per_second": 7.464,
|
| 1014 |
+
"step": 14267
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 17.04,
|
| 1018 |
+
"learning_rate": 3e-05,
|
| 1019 |
+
"loss": 0.6241,
|
| 1020 |
+
"step": 14300
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 17.16,
|
| 1024 |
+
"learning_rate": 3e-05,
|
| 1025 |
+
"loss": 0.5755,
|
| 1026 |
+
"step": 14400
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 17.28,
|
| 1030 |
+
"learning_rate": 3e-05,
|
| 1031 |
+
"loss": 0.6014,
|
| 1032 |
+
"step": 14500
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 17.4,
|
| 1036 |
+
"learning_rate": 3e-05,
|
| 1037 |
+
"loss": 0.5944,
|
| 1038 |
+
"step": 14600
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 17.52,
|
| 1042 |
+
"learning_rate": 3e-05,
|
| 1043 |
+
"loss": 0.6047,
|
| 1044 |
+
"step": 14700
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 17.63,
|
| 1048 |
+
"learning_rate": 3e-05,
|
| 1049 |
+
"loss": 0.5947,
|
| 1050 |
+
"step": 14800
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 17.75,
|
| 1054 |
+
"learning_rate": 3e-05,
|
| 1055 |
+
"loss": 0.6076,
|
| 1056 |
+
"step": 14900
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 17.87,
|
| 1060 |
+
"learning_rate": 3e-05,
|
| 1061 |
+
"loss": 0.6114,
|
| 1062 |
+
"step": 15000
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 17.99,
|
| 1066 |
+
"learning_rate": 3e-05,
|
| 1067 |
+
"loss": 0.6092,
|
| 1068 |
+
"step": 15100
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 18.0,
|
| 1072 |
+
"eval_accuracy": 0.7773701657458564,
|
| 1073 |
+
"eval_loss": 1.0243878364562988,
|
| 1074 |
+
"eval_runtime": 8.362,
|
| 1075 |
+
"eval_samples_per_second": 59.794,
|
| 1076 |
+
"eval_steps_per_second": 7.534,
|
| 1077 |
+
"step": 15106
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 18.11,
|
| 1081 |
+
"learning_rate": 3e-05,
|
| 1082 |
+
"loss": 0.5702,
|
| 1083 |
+
"step": 15200
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 18.23,
|
| 1087 |
+
"learning_rate": 3e-05,
|
| 1088 |
+
"loss": 0.5506,
|
| 1089 |
+
"step": 15300
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 18.35,
|
| 1093 |
+
"learning_rate": 3e-05,
|
| 1094 |
+
"loss": 0.5814,
|
| 1095 |
+
"step": 15400
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 18.47,
|
| 1099 |
+
"learning_rate": 3e-05,
|
| 1100 |
+
"loss": 0.5672,
|
| 1101 |
+
"step": 15500
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 18.59,
|
| 1105 |
+
"learning_rate": 3e-05,
|
| 1106 |
+
"loss": 0.566,
|
| 1107 |
+
"step": 15600
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 18.71,
|
| 1111 |
+
"learning_rate": 3e-05,
|
| 1112 |
+
"loss": 0.571,
|
| 1113 |
+
"step": 15700
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 18.83,
|
| 1117 |
+
"learning_rate": 3e-05,
|
| 1118 |
+
"loss": 0.5795,
|
| 1119 |
+
"step": 15800
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 18.95,
|
| 1123 |
+
"learning_rate": 3e-05,
|
| 1124 |
+
"loss": 0.5946,
|
| 1125 |
+
"step": 15900
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 19.0,
|
| 1129 |
+
"eval_accuracy": 0.7783977900552487,
|
| 1130 |
+
"eval_loss": 1.0291906595230103,
|
| 1131 |
+
"eval_runtime": 8.1655,
|
| 1132 |
+
"eval_samples_per_second": 61.234,
|
| 1133 |
+
"eval_steps_per_second": 7.715,
|
| 1134 |
+
"step": 15945
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 19.06,
|
| 1138 |
+
"learning_rate": 3e-05,
|
| 1139 |
+
"loss": 0.5589,
|
| 1140 |
+
"step": 16000
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 19.18,
|
| 1144 |
+
"learning_rate": 3e-05,
|
| 1145 |
+
"loss": 0.5427,
|
| 1146 |
+
"step": 16100
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 19.3,
|
| 1150 |
+
"learning_rate": 3e-05,
|
| 1151 |
+
"loss": 0.5502,
|
| 1152 |
+
"step": 16200
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 19.42,
|
| 1156 |
+
"learning_rate": 3e-05,
|
| 1157 |
+
"loss": 0.5433,
|
| 1158 |
+
"step": 16300
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 19.54,
|
| 1162 |
+
"learning_rate": 3e-05,
|
| 1163 |
+
"loss": 0.5485,
|
| 1164 |
+
"step": 16400
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 19.66,
|
| 1168 |
+
"learning_rate": 3e-05,
|
| 1169 |
+
"loss": 0.5603,
|
| 1170 |
+
"step": 16500
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 19.78,
|
| 1174 |
+
"learning_rate": 3e-05,
|
| 1175 |
+
"loss": 0.5547,
|
| 1176 |
+
"step": 16600
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 19.9,
|
| 1180 |
+
"learning_rate": 3e-05,
|
| 1181 |
+
"loss": 0.5526,
|
| 1182 |
+
"step": 16700
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 19.99,
|
| 1186 |
+
"eval_accuracy": 0.7797974217311234,
|
| 1187 |
+
"eval_loss": 1.0143389701843262,
|
| 1188 |
+
"eval_runtime": 7.342,
|
| 1189 |
+
"eval_samples_per_second": 68.101,
|
| 1190 |
+
"eval_steps_per_second": 8.581,
|
| 1191 |
+
"step": 16780
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 19.99,
|
| 1195 |
+
"step": 16780,
|
| 1196 |
+
"total_flos": 3.530345004167856e+18,
|
| 1197 |
+
"train_loss": 0.8347494439089824,
|
| 1198 |
+
"train_runtime": 13269.4265,
|
| 1199 |
+
"train_samples_per_second": 40.475,
|
| 1200 |
+
"train_steps_per_second": 1.265
|
| 1201 |
+
}
|
| 1202 |
+
],
|
| 1203 |
+
"logging_steps": 100,
|
| 1204 |
+
"max_steps": 16780,
|
| 1205 |
+
"num_train_epochs": 20,
|
| 1206 |
+
"save_steps": 500,
|
| 1207 |
+
"total_flos": 3.530345004167856e+18,
|
| 1208 |
+
"trial_name": null,
|
| 1209 |
+
"trial_params": null
|
| 1210 |
+
}
|