End of training
Browse files- README.md +14 -2
- all_results.json +12 -12
- eval_results.json +7 -7
- tokenizer.json +1 -6
- train_results.json +6 -6
- trainer_state.json +690 -12
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_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_lora2
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [Qwen/Qwen1.5-4B](https://huggingface.co/Qwen/Qwen1.5-4B) on
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.7825
|
| 22 |
- Accuracy: 0.7891
|
|
|
|
| 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_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: 0.7890842332613391
|
| 23 |
library_name: peft
|
| 24 |
---
|
| 25 |
|
|
|
|
| 28 |
|
| 29 |
# lmind_hotpot_train8000_eval7405_v1_docidx_Qwen_Qwen1.5-4B_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: 0.7825
|
| 34 |
- Accuracy: 0.7891
|
all_results.json
CHANGED
|
@@ -1,16 +1,16 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime": 7.
|
| 6 |
"eval_samples": 500,
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second": 8.
|
| 9 |
-
"perplexity": 2.
|
| 10 |
-
"total_flos":
|
| 11 |
-
"train_loss":
|
| 12 |
-
"train_runtime":
|
| 13 |
"train_samples": 26854,
|
| 14 |
-
"train_samples_per_second":
|
| 15 |
-
"train_steps_per_second": 0.
|
| 16 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 19.997021149836165,
|
| 3 |
+
"eval_accuracy": 0.7890842332613391,
|
| 4 |
+
"eval_loss": 0.7825167179107666,
|
| 5 |
+
"eval_runtime": 7.775,
|
| 6 |
"eval_samples": 500,
|
| 7 |
+
"eval_samples_per_second": 64.309,
|
| 8 |
+
"eval_steps_per_second": 8.103,
|
| 9 |
+
"perplexity": 2.186969330199743,
|
| 10 |
+
"total_flos": 1.3732763132881797e+18,
|
| 11 |
+
"train_loss": 0.18793870911126484,
|
| 12 |
+
"train_runtime": 19785.1606,
|
| 13 |
"train_samples": 26854,
|
| 14 |
+
"train_samples_per_second": 27.146,
|
| 15 |
+
"train_steps_per_second": 0.848
|
| 16 |
}
|
eval_results.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime": 7.
|
| 6 |
"eval_samples": 500,
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second": 8.
|
| 9 |
-
"perplexity": 2.
|
| 10 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 19.997021149836165,
|
| 3 |
+
"eval_accuracy": 0.7890842332613391,
|
| 4 |
+
"eval_loss": 0.7825167179107666,
|
| 5 |
+
"eval_runtime": 7.775,
|
| 6 |
"eval_samples": 500,
|
| 7 |
+
"eval_samples_per_second": 64.309,
|
| 8 |
+
"eval_steps_per_second": 8.103,
|
| 9 |
+
"perplexity": 2.186969330199743
|
| 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
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 26854,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 19.997021149836165,
|
| 3 |
+
"total_flos": 1.3732763132881797e+18,
|
| 4 |
+
"train_loss": 0.18793870911126484,
|
| 5 |
+
"train_runtime": 19785.1606,
|
| 6 |
"train_samples": 26854,
|
| 7 |
+
"train_samples_per_second": 27.146,
|
| 8 |
+
"train_steps_per_second": 0.848
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -680,21 +680,699 @@
|
|
| 680 |
"step": 8390
|
| 681 |
},
|
| 682 |
{
|
| 683 |
-
"epoch":
|
| 684 |
-
"
|
| 685 |
-
"
|
| 686 |
-
"
|
| 687 |
-
"
|
| 688 |
-
|
| 689 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 690 |
}
|
| 691 |
],
|
| 692 |
"logging_steps": 100,
|
| 693 |
-
"max_steps":
|
| 694 |
"num_input_tokens_seen": 0,
|
| 695 |
-
"num_train_epochs":
|
| 696 |
"save_steps": 500,
|
| 697 |
-
"total_flos":
|
| 698 |
"train_batch_size": 1,
|
| 699 |
"trial_name": null,
|
| 700 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 19.997021149836165,
|
| 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,
|
|
|
|
| 680 |
"step": 8390
|
| 681 |
},
|
| 682 |
{
|
| 683 |
+
"epoch": 10.011915400655347,
|
| 684 |
+
"grad_norm": 1.1237155199050903,
|
| 685 |
+
"learning_rate": 0.0001,
|
| 686 |
+
"loss": 0.5067,
|
| 687 |
+
"step": 8400
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 10.131069407208818,
|
| 691 |
+
"grad_norm": 1.2166295051574707,
|
| 692 |
+
"learning_rate": 0.0001,
|
| 693 |
+
"loss": 0.5138,
|
| 694 |
+
"step": 8500
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 10.250223413762289,
|
| 698 |
+
"grad_norm": 1.1621054410934448,
|
| 699 |
+
"learning_rate": 0.0001,
|
| 700 |
+
"loss": 0.5314,
|
| 701 |
+
"step": 8600
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 10.369377420315757,
|
| 705 |
+
"grad_norm": 1.082357406616211,
|
| 706 |
+
"learning_rate": 0.0001,
|
| 707 |
+
"loss": 0.5358,
|
| 708 |
+
"step": 8700
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 10.488531426869228,
|
| 712 |
+
"grad_norm": 1.3300856351852417,
|
| 713 |
+
"learning_rate": 0.0001,
|
| 714 |
+
"loss": 0.5464,
|
| 715 |
+
"step": 8800
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 10.607685433422699,
|
| 719 |
+
"grad_norm": 1.1997504234313965,
|
| 720 |
+
"learning_rate": 0.0001,
|
| 721 |
+
"loss": 0.5422,
|
| 722 |
+
"step": 8900
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 10.72683943997617,
|
| 726 |
+
"grad_norm": 1.4044774770736694,
|
| 727 |
+
"learning_rate": 0.0001,
|
| 728 |
+
"loss": 0.5567,
|
| 729 |
+
"step": 9000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 10.84599344652964,
|
| 733 |
+
"grad_norm": 1.215906023979187,
|
| 734 |
+
"learning_rate": 0.0001,
|
| 735 |
+
"loss": 0.562,
|
| 736 |
+
"step": 9100
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 10.96514745308311,
|
| 740 |
+
"grad_norm": 1.2198461294174194,
|
| 741 |
+
"learning_rate": 0.0001,
|
| 742 |
+
"loss": 0.5784,
|
| 743 |
+
"step": 9200
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 10.999702114983616,
|
| 747 |
+
"eval_accuracy": 0.7730842332613391,
|
| 748 |
+
"eval_loss": 1.010130763053894,
|
| 749 |
+
"eval_runtime": 7.8978,
|
| 750 |
+
"eval_samples_per_second": 63.309,
|
| 751 |
+
"eval_steps_per_second": 7.977,
|
| 752 |
+
"step": 9229
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 11.08430145963658,
|
| 756 |
+
"grad_norm": 1.3466765880584717,
|
| 757 |
+
"learning_rate": 0.0001,
|
| 758 |
+
"loss": 0.4876,
|
| 759 |
+
"step": 9300
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 11.20345546619005,
|
| 763 |
+
"grad_norm": 1.3371703624725342,
|
| 764 |
+
"learning_rate": 0.0001,
|
| 765 |
+
"loss": 0.4666,
|
| 766 |
+
"step": 9400
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 11.32260947274352,
|
| 770 |
+
"grad_norm": 1.3903799057006836,
|
| 771 |
+
"learning_rate": 0.0001,
|
| 772 |
+
"loss": 0.4819,
|
| 773 |
+
"step": 9500
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 11.441763479296991,
|
| 777 |
+
"grad_norm": 1.261116623878479,
|
| 778 |
+
"learning_rate": 0.0001,
|
| 779 |
+
"loss": 0.4857,
|
| 780 |
+
"step": 9600
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 11.560917485850462,
|
| 784 |
+
"grad_norm": 1.4429560899734497,
|
| 785 |
+
"learning_rate": 0.0001,
|
| 786 |
+
"loss": 0.4941,
|
| 787 |
+
"step": 9700
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 11.680071492403933,
|
| 791 |
+
"grad_norm": 1.268157958984375,
|
| 792 |
+
"learning_rate": 0.0001,
|
| 793 |
+
"loss": 0.4913,
|
| 794 |
+
"step": 9800
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 11.799225498957403,
|
| 798 |
+
"grad_norm": 1.3515466451644897,
|
| 799 |
+
"learning_rate": 0.0001,
|
| 800 |
+
"loss": 0.5074,
|
| 801 |
+
"step": 9900
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 11.918379505510872,
|
| 805 |
+
"grad_norm": 1.343897819519043,
|
| 806 |
+
"learning_rate": 0.0001,
|
| 807 |
+
"loss": 0.5071,
|
| 808 |
+
"step": 10000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 11.999404229967233,
|
| 812 |
+
"eval_accuracy": 0.7760388768898488,
|
| 813 |
+
"eval_loss": 0.9537739753723145,
|
| 814 |
+
"eval_runtime": 7.6808,
|
| 815 |
+
"eval_samples_per_second": 65.098,
|
| 816 |
+
"eval_steps_per_second": 8.202,
|
| 817 |
+
"step": 10068
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 12.037533512064343,
|
| 821 |
+
"grad_norm": 1.2503083944320679,
|
| 822 |
+
"learning_rate": 0.0001,
|
| 823 |
+
"loss": 0.4805,
|
| 824 |
+
"step": 10100
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 12.156687518617813,
|
| 828 |
+
"grad_norm": 1.1671448945999146,
|
| 829 |
+
"learning_rate": 0.0001,
|
| 830 |
+
"loss": 0.4098,
|
| 831 |
+
"step": 10200
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 12.275841525171284,
|
| 835 |
+
"grad_norm": 1.2349199056625366,
|
| 836 |
+
"learning_rate": 0.0001,
|
| 837 |
+
"loss": 0.4222,
|
| 838 |
+
"step": 10300
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 12.394995531724755,
|
| 842 |
+
"grad_norm": 1.3856853246688843,
|
| 843 |
+
"learning_rate": 0.0001,
|
| 844 |
+
"loss": 0.4282,
|
| 845 |
+
"step": 10400
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 12.514149538278225,
|
| 849 |
+
"grad_norm": 1.2162753343582153,
|
| 850 |
+
"learning_rate": 0.0001,
|
| 851 |
+
"loss": 0.4353,
|
| 852 |
+
"step": 10500
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 12.633303544831694,
|
| 856 |
+
"grad_norm": 1.4368740320205688,
|
| 857 |
+
"learning_rate": 0.0001,
|
| 858 |
+
"loss": 0.4506,
|
| 859 |
+
"step": 10600
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 12.752457551385165,
|
| 863 |
+
"grad_norm": 1.2878433465957642,
|
| 864 |
+
"learning_rate": 0.0001,
|
| 865 |
+
"loss": 0.4607,
|
| 866 |
+
"step": 10700
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 12.871611557938635,
|
| 870 |
+
"grad_norm": 1.3356980085372925,
|
| 871 |
+
"learning_rate": 0.0001,
|
| 872 |
+
"loss": 0.4651,
|
| 873 |
+
"step": 10800
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 12.990765564492106,
|
| 877 |
+
"grad_norm": 1.4646358489990234,
|
| 878 |
+
"learning_rate": 0.0001,
|
| 879 |
+
"loss": 0.4734,
|
| 880 |
+
"step": 10900
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 12.99910634495085,
|
| 884 |
+
"eval_accuracy": 0.7790539956803456,
|
| 885 |
+
"eval_loss": 0.9292365908622742,
|
| 886 |
+
"eval_runtime": 7.7765,
|
| 887 |
+
"eval_samples_per_second": 64.296,
|
| 888 |
+
"eval_steps_per_second": 8.101,
|
| 889 |
+
"step": 10907
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 13.109919571045577,
|
| 893 |
+
"grad_norm": 1.0571186542510986,
|
| 894 |
+
"learning_rate": 0.0001,
|
| 895 |
+
"loss": 0.3875,
|
| 896 |
+
"step": 11000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 13.229073577599047,
|
| 900 |
+
"grad_norm": 1.5074485540390015,
|
| 901 |
+
"learning_rate": 0.0001,
|
| 902 |
+
"loss": 0.3734,
|
| 903 |
+
"step": 11100
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 13.348227584152518,
|
| 907 |
+
"grad_norm": 1.0904532670974731,
|
| 908 |
+
"learning_rate": 0.0001,
|
| 909 |
+
"loss": 0.3906,
|
| 910 |
+
"step": 11200
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 13.467381590705987,
|
| 914 |
+
"grad_norm": 1.1496778726577759,
|
| 915 |
+
"learning_rate": 0.0001,
|
| 916 |
+
"loss": 0.3999,
|
| 917 |
+
"step": 11300
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 13.586535597259457,
|
| 921 |
+
"grad_norm": 1.4542232751846313,
|
| 922 |
+
"learning_rate": 0.0001,
|
| 923 |
+
"loss": 0.4048,
|
| 924 |
+
"step": 11400
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 13.705689603812928,
|
| 928 |
+
"grad_norm": 1.6107604503631592,
|
| 929 |
+
"learning_rate": 0.0001,
|
| 930 |
+
"loss": 0.4098,
|
| 931 |
+
"step": 11500
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 13.824843610366399,
|
| 935 |
+
"grad_norm": 1.6126165390014648,
|
| 936 |
+
"learning_rate": 0.0001,
|
| 937 |
+
"loss": 0.4141,
|
| 938 |
+
"step": 11600
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 13.94399761691987,
|
| 942 |
+
"grad_norm": 1.5488628149032593,
|
| 943 |
+
"learning_rate": 0.0001,
|
| 944 |
+
"loss": 0.4302,
|
| 945 |
+
"step": 11700
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 14.0,
|
| 949 |
+
"eval_accuracy": 0.7808639308855292,
|
| 950 |
+
"eval_loss": 0.8845791220664978,
|
| 951 |
+
"eval_runtime": 7.816,
|
| 952 |
+
"eval_samples_per_second": 63.971,
|
| 953 |
+
"eval_steps_per_second": 8.06,
|
| 954 |
+
"step": 11747
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 14.06315162347334,
|
| 958 |
+
"grad_norm": 1.1683179140090942,
|
| 959 |
+
"learning_rate": 0.0001,
|
| 960 |
+
"loss": 0.3842,
|
| 961 |
+
"step": 11800
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 14.182305630026809,
|
| 965 |
+
"grad_norm": 1.4228167533874512,
|
| 966 |
+
"learning_rate": 0.0001,
|
| 967 |
+
"loss": 0.3452,
|
| 968 |
+
"step": 11900
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 14.30145963658028,
|
| 972 |
+
"grad_norm": 1.3090318441390991,
|
| 973 |
+
"learning_rate": 0.0001,
|
| 974 |
+
"loss": 0.3518,
|
| 975 |
+
"step": 12000
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 14.42061364313375,
|
| 979 |
+
"grad_norm": 1.3104828596115112,
|
| 980 |
+
"learning_rate": 0.0001,
|
| 981 |
+
"loss": 0.3632,
|
| 982 |
+
"step": 12100
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 14.53976764968722,
|
| 986 |
+
"grad_norm": 1.5828286409378052,
|
| 987 |
+
"learning_rate": 0.0001,
|
| 988 |
+
"loss": 0.3737,
|
| 989 |
+
"step": 12200
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 14.658921656240691,
|
| 993 |
+
"grad_norm": 1.3281217813491821,
|
| 994 |
+
"learning_rate": 0.0001,
|
| 995 |
+
"loss": 0.3705,
|
| 996 |
+
"step": 12300
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 14.778075662794162,
|
| 1000 |
+
"grad_norm": 1.753090500831604,
|
| 1001 |
+
"learning_rate": 0.0001,
|
| 1002 |
+
"loss": 0.3839,
|
| 1003 |
+
"step": 12400
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 14.897229669347631,
|
| 1007 |
+
"grad_norm": 1.3213167190551758,
|
| 1008 |
+
"learning_rate": 0.0001,
|
| 1009 |
+
"loss": 0.3917,
|
| 1010 |
+
"step": 12500
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 14.999702114983616,
|
| 1014 |
+
"eval_accuracy": 0.7832613390928725,
|
| 1015 |
+
"eval_loss": 0.8536106944084167,
|
| 1016 |
+
"eval_runtime": 7.6693,
|
| 1017 |
+
"eval_samples_per_second": 65.195,
|
| 1018 |
+
"eval_steps_per_second": 8.215,
|
| 1019 |
+
"step": 12586
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 15.016383675901102,
|
| 1023 |
+
"grad_norm": 1.5647796392440796,
|
| 1024 |
+
"learning_rate": 0.0001,
|
| 1025 |
+
"loss": 0.3796,
|
| 1026 |
+
"step": 12600
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 15.135537682454572,
|
| 1030 |
+
"grad_norm": 1.903051495552063,
|
| 1031 |
+
"learning_rate": 0.0001,
|
| 1032 |
+
"loss": 0.3098,
|
| 1033 |
+
"step": 12700
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 15.254691689008043,
|
| 1037 |
+
"grad_norm": 1.4900842905044556,
|
| 1038 |
+
"learning_rate": 0.0001,
|
| 1039 |
+
"loss": 0.3291,
|
| 1040 |
+
"step": 12800
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 15.373845695561513,
|
| 1044 |
+
"grad_norm": 1.3253552913665771,
|
| 1045 |
+
"learning_rate": 0.0001,
|
| 1046 |
+
"loss": 0.3284,
|
| 1047 |
+
"step": 12900
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 15.492999702114984,
|
| 1051 |
+
"grad_norm": 1.582220196723938,
|
| 1052 |
+
"learning_rate": 0.0001,
|
| 1053 |
+
"loss": 0.345,
|
| 1054 |
+
"step": 13000
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 15.612153708668455,
|
| 1058 |
+
"grad_norm": 1.4743067026138306,
|
| 1059 |
+
"learning_rate": 0.0001,
|
| 1060 |
+
"loss": 0.3482,
|
| 1061 |
+
"step": 13100
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 15.731307715221924,
|
| 1065 |
+
"grad_norm": 1.5687114000320435,
|
| 1066 |
+
"learning_rate": 0.0001,
|
| 1067 |
+
"loss": 0.3573,
|
| 1068 |
+
"step": 13200
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 15.850461721775394,
|
| 1072 |
+
"grad_norm": 1.5427637100219727,
|
| 1073 |
+
"learning_rate": 0.0001,
|
| 1074 |
+
"loss": 0.3531,
|
| 1075 |
+
"step": 13300
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 15.969615728328865,
|
| 1079 |
+
"grad_norm": 1.621741533279419,
|
| 1080 |
+
"learning_rate": 0.0001,
|
| 1081 |
+
"loss": 0.3632,
|
| 1082 |
+
"step": 13400
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 15.999404229967233,
|
| 1086 |
+
"eval_accuracy": 0.7846133909287257,
|
| 1087 |
+
"eval_loss": 0.8468108177185059,
|
| 1088 |
+
"eval_runtime": 7.8749,
|
| 1089 |
+
"eval_samples_per_second": 63.493,
|
| 1090 |
+
"eval_steps_per_second": 8.0,
|
| 1091 |
+
"step": 13425
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 16.088769734882334,
|
| 1095 |
+
"grad_norm": 1.3480048179626465,
|
| 1096 |
+
"learning_rate": 0.0001,
|
| 1097 |
+
"loss": 0.3116,
|
| 1098 |
+
"step": 13500
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 16.207923741435806,
|
| 1102 |
+
"grad_norm": 1.3218774795532227,
|
| 1103 |
+
"learning_rate": 0.0001,
|
| 1104 |
+
"loss": 0.2957,
|
| 1105 |
+
"step": 13600
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 16.327077747989275,
|
| 1109 |
+
"grad_norm": 1.5867496728897095,
|
| 1110 |
+
"learning_rate": 0.0001,
|
| 1111 |
+
"loss": 0.3087,
|
| 1112 |
+
"step": 13700
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 16.446231754542747,
|
| 1116 |
+
"grad_norm": 1.3426684141159058,
|
| 1117 |
+
"learning_rate": 0.0001,
|
| 1118 |
+
"loss": 0.315,
|
| 1119 |
+
"step": 13800
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 16.565385761096216,
|
| 1123 |
+
"grad_norm": 1.5667626857757568,
|
| 1124 |
+
"learning_rate": 0.0001,
|
| 1125 |
+
"loss": 0.3212,
|
| 1126 |
+
"step": 13900
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 16.68453976764969,
|
| 1130 |
+
"grad_norm": 1.3792177438735962,
|
| 1131 |
+
"learning_rate": 0.0001,
|
| 1132 |
+
"loss": 0.3258,
|
| 1133 |
+
"step": 14000
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 16.803693774203158,
|
| 1137 |
+
"grad_norm": 1.5244312286376953,
|
| 1138 |
+
"learning_rate": 0.0001,
|
| 1139 |
+
"loss": 0.3337,
|
| 1140 |
+
"step": 14100
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 16.922847780756626,
|
| 1144 |
+
"grad_norm": 1.598897933959961,
|
| 1145 |
+
"learning_rate": 0.0001,
|
| 1146 |
+
"loss": 0.3351,
|
| 1147 |
+
"step": 14200
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 16.99910634495085,
|
| 1151 |
+
"eval_accuracy": 0.7862505399568035,
|
| 1152 |
+
"eval_loss": 0.8243688941001892,
|
| 1153 |
+
"eval_runtime": 7.7339,
|
| 1154 |
+
"eval_samples_per_second": 64.651,
|
| 1155 |
+
"eval_steps_per_second": 8.146,
|
| 1156 |
+
"step": 14264
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 17.0420017873101,
|
| 1160 |
+
"grad_norm": 1.5057127475738525,
|
| 1161 |
+
"learning_rate": 0.0001,
|
| 1162 |
+
"loss": 0.3195,
|
| 1163 |
+
"step": 14300
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 17.161155793863568,
|
| 1167 |
+
"grad_norm": 1.175302505493164,
|
| 1168 |
+
"learning_rate": 0.0001,
|
| 1169 |
+
"loss": 0.2775,
|
| 1170 |
+
"step": 14400
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 17.28030980041704,
|
| 1174 |
+
"grad_norm": 1.4136508703231812,
|
| 1175 |
+
"learning_rate": 0.0001,
|
| 1176 |
+
"loss": 0.289,
|
| 1177 |
+
"step": 14500
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 17.39946380697051,
|
| 1181 |
+
"grad_norm": 1.5237888097763062,
|
| 1182 |
+
"learning_rate": 0.0001,
|
| 1183 |
+
"loss": 0.2897,
|
| 1184 |
+
"step": 14600
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 17.51861781352398,
|
| 1188 |
+
"grad_norm": 1.4952021837234497,
|
| 1189 |
+
"learning_rate": 0.0001,
|
| 1190 |
+
"loss": 0.3002,
|
| 1191 |
+
"step": 14700
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 17.63777182007745,
|
| 1195 |
+
"grad_norm": 1.4359500408172607,
|
| 1196 |
+
"learning_rate": 0.0001,
|
| 1197 |
+
"loss": 0.2994,
|
| 1198 |
+
"step": 14800
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 17.75692582663092,
|
| 1202 |
+
"grad_norm": 1.2760846614837646,
|
| 1203 |
+
"learning_rate": 0.0001,
|
| 1204 |
+
"loss": 0.3085,
|
| 1205 |
+
"step": 14900
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 17.87607983318439,
|
| 1209 |
+
"grad_norm": 1.4437624216079712,
|
| 1210 |
+
"learning_rate": 0.0001,
|
| 1211 |
+
"loss": 0.3141,
|
| 1212 |
+
"step": 15000
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 17.99523383973786,
|
| 1216 |
+
"grad_norm": 1.4819003343582153,
|
| 1217 |
+
"learning_rate": 0.0001,
|
| 1218 |
+
"loss": 0.3186,
|
| 1219 |
+
"step": 15100
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 18.0,
|
| 1223 |
+
"eval_accuracy": 0.7870842332613391,
|
| 1224 |
+
"eval_loss": 0.8095716834068298,
|
| 1225 |
+
"eval_runtime": 7.8907,
|
| 1226 |
+
"eval_samples_per_second": 63.366,
|
| 1227 |
+
"eval_steps_per_second": 7.984,
|
| 1228 |
+
"step": 15104
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 18.114387846291333,
|
| 1232 |
+
"grad_norm": 1.4048779010772705,
|
| 1233 |
+
"learning_rate": 0.0001,
|
| 1234 |
+
"loss": 0.2551,
|
| 1235 |
+
"step": 15200
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 18.2335418528448,
|
| 1239 |
+
"grad_norm": 1.2755182981491089,
|
| 1240 |
+
"learning_rate": 0.0001,
|
| 1241 |
+
"loss": 0.2625,
|
| 1242 |
+
"step": 15300
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 18.35269585939827,
|
| 1246 |
+
"grad_norm": 1.5804539918899536,
|
| 1247 |
+
"learning_rate": 0.0001,
|
| 1248 |
+
"loss": 0.2733,
|
| 1249 |
+
"step": 15400
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 18.471849865951743,
|
| 1253 |
+
"grad_norm": 1.2414181232452393,
|
| 1254 |
+
"learning_rate": 0.0001,
|
| 1255 |
+
"loss": 0.2827,
|
| 1256 |
+
"step": 15500
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 18.591003872505212,
|
| 1260 |
+
"grad_norm": 1.2644073963165283,
|
| 1261 |
+
"learning_rate": 0.0001,
|
| 1262 |
+
"loss": 0.2873,
|
| 1263 |
+
"step": 15600
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 18.710157879058684,
|
| 1267 |
+
"grad_norm": 1.8947105407714844,
|
| 1268 |
+
"learning_rate": 0.0001,
|
| 1269 |
+
"loss": 0.2913,
|
| 1270 |
+
"step": 15700
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 18.829311885612153,
|
| 1274 |
+
"grad_norm": 1.429527759552002,
|
| 1275 |
+
"learning_rate": 0.0001,
|
| 1276 |
+
"loss": 0.2977,
|
| 1277 |
+
"step": 15800
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 18.948465892165625,
|
| 1281 |
+
"grad_norm": 1.8058485984802246,
|
| 1282 |
+
"learning_rate": 0.0001,
|
| 1283 |
+
"loss": 0.2957,
|
| 1284 |
+
"step": 15900
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 18.999702114983616,
|
| 1288 |
+
"eval_accuracy": 0.7885053995680346,
|
| 1289 |
+
"eval_loss": 0.7864968776702881,
|
| 1290 |
+
"eval_runtime": 7.7324,
|
| 1291 |
+
"eval_samples_per_second": 64.663,
|
| 1292 |
+
"eval_steps_per_second": 8.148,
|
| 1293 |
+
"step": 15943
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 19.067619898719094,
|
| 1297 |
+
"grad_norm": 1.3445817232131958,
|
| 1298 |
+
"learning_rate": 0.0001,
|
| 1299 |
+
"loss": 0.2748,
|
| 1300 |
+
"step": 16000
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 19.186773905272563,
|
| 1304 |
+
"grad_norm": 1.358742356300354,
|
| 1305 |
+
"learning_rate": 0.0001,
|
| 1306 |
+
"loss": 0.2494,
|
| 1307 |
+
"step": 16100
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 19.305927911826036,
|
| 1311 |
+
"grad_norm": 1.2323551177978516,
|
| 1312 |
+
"learning_rate": 0.0001,
|
| 1313 |
+
"loss": 0.255,
|
| 1314 |
+
"step": 16200
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 19.425081918379504,
|
| 1318 |
+
"grad_norm": 1.210010290145874,
|
| 1319 |
+
"learning_rate": 0.0001,
|
| 1320 |
+
"loss": 0.2592,
|
| 1321 |
+
"step": 16300
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 19.544235924932977,
|
| 1325 |
+
"grad_norm": 1.4125585556030273,
|
| 1326 |
+
"learning_rate": 0.0001,
|
| 1327 |
+
"loss": 0.2664,
|
| 1328 |
+
"step": 16400
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 19.663389931486446,
|
| 1332 |
+
"grad_norm": 1.5337769985198975,
|
| 1333 |
+
"learning_rate": 0.0001,
|
| 1334 |
+
"loss": 0.2741,
|
| 1335 |
+
"step": 16500
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 19.782543938039918,
|
| 1339 |
+
"grad_norm": 1.4104398488998413,
|
| 1340 |
+
"learning_rate": 0.0001,
|
| 1341 |
+
"loss": 0.277,
|
| 1342 |
+
"step": 16600
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 19.901697944593387,
|
| 1346 |
+
"grad_norm": 1.6847435235977173,
|
| 1347 |
+
"learning_rate": 0.0001,
|
| 1348 |
+
"loss": 0.2858,
|
| 1349 |
+
"step": 16700
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 19.997021149836165,
|
| 1353 |
+
"eval_accuracy": 0.7890842332613391,
|
| 1354 |
+
"eval_loss": 0.7825167179107666,
|
| 1355 |
+
"eval_runtime": 7.7619,
|
| 1356 |
+
"eval_samples_per_second": 64.417,
|
| 1357 |
+
"eval_steps_per_second": 8.117,
|
| 1358 |
+
"step": 16780
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 19.997021149836165,
|
| 1362 |
+
"step": 16780,
|
| 1363 |
+
"total_flos": 1.3732763132881797e+18,
|
| 1364 |
+
"train_loss": 0.18793870911126484,
|
| 1365 |
+
"train_runtime": 19785.1606,
|
| 1366 |
+
"train_samples_per_second": 27.146,
|
| 1367 |
+
"train_steps_per_second": 0.848
|
| 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": 1.3732763132881797e+18,
|
| 1376 |
"train_batch_size": 1,
|
| 1377 |
"trial_name": null,
|
| 1378 |
"trial_params": null
|