End of training
Browse files- README.md +4 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +9 -0
- trainer_state.json +2179 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: apache-2.0
|
|
| 4 |
base_model: AdaptLLM/biomed-Qwen2-VL-2B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: qwenvl-2B-cadica-direction-scale4
|
|
@@ -15,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# qwenvl-2B-cadica-direction-scale4
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [AdaptLLM/biomed-Qwen2-VL-2B-Instruct](https://huggingface.co/AdaptLLM/biomed-Qwen2-VL-2B-Instruct) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
-
- Num Input Tokens Seen:
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 4 |
base_model: AdaptLLM/biomed-Qwen2-VL-2B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- lora
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: qwenvl-2B-cadica-direction-scale4
|
|
|
|
| 16 |
|
| 17 |
# qwenvl-2B-cadica-direction-scale4
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [AdaptLLM/biomed-Qwen2-VL-2B-Instruct](https://huggingface.co/AdaptLLM/biomed-Qwen2-VL-2B-Instruct) on the CADICA血管分支方向題scale4(TRAIN) dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.0000
|
| 22 |
+
- Num Input Tokens Seen: 11990784
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.6185938707185166,
|
| 3 |
+
"eval_loss": 1.5566551155643538e-05,
|
| 4 |
+
"eval_runtime": 46.6857,
|
| 5 |
+
"eval_samples_per_second": 1.285,
|
| 6 |
+
"eval_steps_per_second": 0.321,
|
| 7 |
+
"num_input_tokens_seen": 11990784,
|
| 8 |
+
"total_flos": 791107965943808.0,
|
| 9 |
+
"train_loss": 6.11944748344618e-08,
|
| 10 |
+
"train_runtime": 102.3684,
|
| 11 |
+
"train_samples_per_second": 281.337,
|
| 12 |
+
"train_steps_per_second": 11.722
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.6185938707185166,
|
| 3 |
+
"eval_loss": 1.5566551155643538e-05,
|
| 4 |
+
"eval_runtime": 46.6857,
|
| 5 |
+
"eval_samples_per_second": 1.285,
|
| 6 |
+
"eval_steps_per_second": 0.321,
|
| 7 |
+
"num_input_tokens_seen": 11990784
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.6185938707185166,
|
| 3 |
+
"num_input_tokens_seen": 11990784,
|
| 4 |
+
"total_flos": 791107965943808.0,
|
| 5 |
+
"train_loss": 6.11944748344618e-08,
|
| 6 |
+
"train_runtime": 102.3684,
|
| 7 |
+
"train_samples_per_second": 281.337,
|
| 8 |
+
"train_steps_per_second": 11.722
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2179 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.5566551155643538e-05,
|
| 3 |
+
"best_model_checkpoint": "saves/CADICA_qwenvl_direction_scale4/lora/sft/checkpoint-1150",
|
| 4 |
+
"epoch": 0.6185938707185166,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 1201,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0025753283543651817,
|
| 13 |
+
"grad_norm": 14.230540018977528,
|
| 14 |
+
"learning_rate": 2.9411764705882355e-06,
|
| 15 |
+
"loss": 1.0954,
|
| 16 |
+
"num_input_tokens_seen": 49920,
|
| 17 |
+
"step": 5
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0051506567087303634,
|
| 21 |
+
"grad_norm": 12.725175128208063,
|
| 22 |
+
"learning_rate": 5.882352941176471e-06,
|
| 23 |
+
"loss": 0.9793,
|
| 24 |
+
"num_input_tokens_seen": 99840,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.007725985063095545,
|
| 29 |
+
"grad_norm": 11.908760913870685,
|
| 30 |
+
"learning_rate": 8.823529411764707e-06,
|
| 31 |
+
"loss": 1.0964,
|
| 32 |
+
"num_input_tokens_seen": 149760,
|
| 33 |
+
"step": 15
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.010301313417460727,
|
| 37 |
+
"grad_norm": 7.5714179545602835,
|
| 38 |
+
"learning_rate": 1.1764705882352942e-05,
|
| 39 |
+
"loss": 0.7079,
|
| 40 |
+
"num_input_tokens_seen": 199680,
|
| 41 |
+
"step": 20
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.012876641771825908,
|
| 45 |
+
"grad_norm": 2.273110551179123,
|
| 46 |
+
"learning_rate": 1.4705882352941177e-05,
|
| 47 |
+
"loss": 0.4213,
|
| 48 |
+
"num_input_tokens_seen": 249600,
|
| 49 |
+
"step": 25
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.01545197012619109,
|
| 53 |
+
"grad_norm": 1.4511084067844011,
|
| 54 |
+
"learning_rate": 1.7647058823529414e-05,
|
| 55 |
+
"loss": 0.3359,
|
| 56 |
+
"num_input_tokens_seen": 299520,
|
| 57 |
+
"step": 30
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.018027298480556272,
|
| 61 |
+
"grad_norm": 1.7462649262033438,
|
| 62 |
+
"learning_rate": 2.058823529411765e-05,
|
| 63 |
+
"loss": 0.3463,
|
| 64 |
+
"num_input_tokens_seen": 349440,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.020602626834921454,
|
| 69 |
+
"grad_norm": 1.2143595820366577,
|
| 70 |
+
"learning_rate": 2.3529411764705884e-05,
|
| 71 |
+
"loss": 0.28,
|
| 72 |
+
"num_input_tokens_seen": 399360,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.023177955189286635,
|
| 77 |
+
"grad_norm": 0.8468719125045373,
|
| 78 |
+
"learning_rate": 2.647058823529412e-05,
|
| 79 |
+
"loss": 0.3442,
|
| 80 |
+
"num_input_tokens_seen": 449280,
|
| 81 |
+
"step": 45
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.025753283543651816,
|
| 85 |
+
"grad_norm": 0.9726867932660042,
|
| 86 |
+
"learning_rate": 2.9411764705882354e-05,
|
| 87 |
+
"loss": 0.3441,
|
| 88 |
+
"num_input_tokens_seen": 499200,
|
| 89 |
+
"step": 50
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.025753283543651816,
|
| 93 |
+
"eval_loss": 0.33834776282310486,
|
| 94 |
+
"eval_runtime": 47.4467,
|
| 95 |
+
"eval_samples_per_second": 1.265,
|
| 96 |
+
"eval_steps_per_second": 0.316,
|
| 97 |
+
"num_input_tokens_seen": 499200,
|
| 98 |
+
"step": 50
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.028328611898016998,
|
| 102 |
+
"grad_norm": 1.4016556961184263,
|
| 103 |
+
"learning_rate": 3.235294117647059e-05,
|
| 104 |
+
"loss": 0.3182,
|
| 105 |
+
"num_input_tokens_seen": 549120,
|
| 106 |
+
"step": 55
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.03090394025238218,
|
| 110 |
+
"grad_norm": 0.6437613769459606,
|
| 111 |
+
"learning_rate": 3.529411764705883e-05,
|
| 112 |
+
"loss": 0.3294,
|
| 113 |
+
"num_input_tokens_seen": 599040,
|
| 114 |
+
"step": 60
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.03347926860674736,
|
| 118 |
+
"grad_norm": 0.7389008951321312,
|
| 119 |
+
"learning_rate": 3.8235294117647055e-05,
|
| 120 |
+
"loss": 0.3097,
|
| 121 |
+
"num_input_tokens_seen": 648960,
|
| 122 |
+
"step": 65
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.036054596961112545,
|
| 126 |
+
"grad_norm": 0.771553860801019,
|
| 127 |
+
"learning_rate": 4.11764705882353e-05,
|
| 128 |
+
"loss": 0.3008,
|
| 129 |
+
"num_input_tokens_seen": 698880,
|
| 130 |
+
"step": 70
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.03862992531547772,
|
| 134 |
+
"grad_norm": 0.6965369148334918,
|
| 135 |
+
"learning_rate": 4.411764705882353e-05,
|
| 136 |
+
"loss": 0.3278,
|
| 137 |
+
"num_input_tokens_seen": 748800,
|
| 138 |
+
"step": 75
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.04120525366984291,
|
| 142 |
+
"grad_norm": 0.912943461315541,
|
| 143 |
+
"learning_rate": 4.705882352941177e-05,
|
| 144 |
+
"loss": 0.3074,
|
| 145 |
+
"num_input_tokens_seen": 798720,
|
| 146 |
+
"step": 80
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.043780582024208085,
|
| 150 |
+
"grad_norm": 0.8407481737577445,
|
| 151 |
+
"learning_rate": 5e-05,
|
| 152 |
+
"loss": 0.3423,
|
| 153 |
+
"num_input_tokens_seen": 848640,
|
| 154 |
+
"step": 85
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.04635591037857327,
|
| 158 |
+
"grad_norm": 0.9112879058417015,
|
| 159 |
+
"learning_rate": 5.294117647058824e-05,
|
| 160 |
+
"loss": 0.3008,
|
| 161 |
+
"num_input_tokens_seen": 898560,
|
| 162 |
+
"step": 90
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 0.04893123873293845,
|
| 166 |
+
"grad_norm": 2.391489040464162,
|
| 167 |
+
"learning_rate": 5.588235294117647e-05,
|
| 168 |
+
"loss": 0.2815,
|
| 169 |
+
"num_input_tokens_seen": 948480,
|
| 170 |
+
"step": 95
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.05150656708730363,
|
| 174 |
+
"grad_norm": 2.155211791607199,
|
| 175 |
+
"learning_rate": 5.882352941176471e-05,
|
| 176 |
+
"loss": 0.2274,
|
| 177 |
+
"num_input_tokens_seen": 998400,
|
| 178 |
+
"step": 100
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.05150656708730363,
|
| 182 |
+
"eval_loss": 0.18663176894187927,
|
| 183 |
+
"eval_runtime": 18.9199,
|
| 184 |
+
"eval_samples_per_second": 3.171,
|
| 185 |
+
"eval_steps_per_second": 0.793,
|
| 186 |
+
"num_input_tokens_seen": 998400,
|
| 187 |
+
"step": 100
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.05408189544166881,
|
| 191 |
+
"grad_norm": 2.2181531422996716,
|
| 192 |
+
"learning_rate": 6.176470588235295e-05,
|
| 193 |
+
"loss": 0.168,
|
| 194 |
+
"num_input_tokens_seen": 1048320,
|
| 195 |
+
"step": 105
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.056657223796033995,
|
| 199 |
+
"grad_norm": 3.1829920225573236,
|
| 200 |
+
"learning_rate": 6.470588235294118e-05,
|
| 201 |
+
"loss": 0.0709,
|
| 202 |
+
"num_input_tokens_seen": 1098240,
|
| 203 |
+
"step": 110
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 0.05923255215039917,
|
| 207 |
+
"grad_norm": 4.337350477588576,
|
| 208 |
+
"learning_rate": 6.764705882352942e-05,
|
| 209 |
+
"loss": 0.1609,
|
| 210 |
+
"num_input_tokens_seen": 1148160,
|
| 211 |
+
"step": 115
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.06180788050476436,
|
| 215 |
+
"grad_norm": 2.1010046045637365,
|
| 216 |
+
"learning_rate": 7.058823529411765e-05,
|
| 217 |
+
"loss": 0.0354,
|
| 218 |
+
"num_input_tokens_seen": 1198080,
|
| 219 |
+
"step": 120
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.06438320885912954,
|
| 223 |
+
"grad_norm": 2.232308844812103,
|
| 224 |
+
"learning_rate": 7.352941176470589e-05,
|
| 225 |
+
"loss": 0.1133,
|
| 226 |
+
"num_input_tokens_seen": 1248000,
|
| 227 |
+
"step": 125
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.06695853721349472,
|
| 231 |
+
"grad_norm": 5.641631090993415,
|
| 232 |
+
"learning_rate": 7.647058823529411e-05,
|
| 233 |
+
"loss": 0.0867,
|
| 234 |
+
"num_input_tokens_seen": 1297920,
|
| 235 |
+
"step": 130
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.0695338655678599,
|
| 239 |
+
"grad_norm": 1.5031437609685787,
|
| 240 |
+
"learning_rate": 7.941176470588235e-05,
|
| 241 |
+
"loss": 0.1352,
|
| 242 |
+
"num_input_tokens_seen": 1347840,
|
| 243 |
+
"step": 135
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.07210919392222509,
|
| 247 |
+
"grad_norm": 3.2992644431188465,
|
| 248 |
+
"learning_rate": 8.23529411764706e-05,
|
| 249 |
+
"loss": 0.101,
|
| 250 |
+
"num_input_tokens_seen": 1397760,
|
| 251 |
+
"step": 140
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.07468452227659027,
|
| 255 |
+
"grad_norm": 3.494236832758233,
|
| 256 |
+
"learning_rate": 8.529411764705883e-05,
|
| 257 |
+
"loss": 0.0334,
|
| 258 |
+
"num_input_tokens_seen": 1447680,
|
| 259 |
+
"step": 145
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.07725985063095545,
|
| 263 |
+
"grad_norm": 0.0602113869322109,
|
| 264 |
+
"learning_rate": 8.823529411764706e-05,
|
| 265 |
+
"loss": 0.0667,
|
| 266 |
+
"num_input_tokens_seen": 1497600,
|
| 267 |
+
"step": 150
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.07725985063095545,
|
| 271 |
+
"eval_loss": 0.09665286540985107,
|
| 272 |
+
"eval_runtime": 19.2745,
|
| 273 |
+
"eval_samples_per_second": 3.113,
|
| 274 |
+
"eval_steps_per_second": 0.778,
|
| 275 |
+
"num_input_tokens_seen": 1497600,
|
| 276 |
+
"step": 150
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.07983517898532062,
|
| 280 |
+
"grad_norm": 6.096163706748617,
|
| 281 |
+
"learning_rate": 9.11764705882353e-05,
|
| 282 |
+
"loss": 0.06,
|
| 283 |
+
"num_input_tokens_seen": 1547520,
|
| 284 |
+
"step": 155
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.08241050733968582,
|
| 288 |
+
"grad_norm": 4.278069142242893,
|
| 289 |
+
"learning_rate": 9.411764705882353e-05,
|
| 290 |
+
"loss": 0.0265,
|
| 291 |
+
"num_input_tokens_seen": 1597440,
|
| 292 |
+
"step": 160
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.08498583569405099,
|
| 296 |
+
"grad_norm": 1.7183097652953412,
|
| 297 |
+
"learning_rate": 9.705882352941177e-05,
|
| 298 |
+
"loss": 0.1743,
|
| 299 |
+
"num_input_tokens_seen": 1647360,
|
| 300 |
+
"step": 165
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 0.08756116404841617,
|
| 304 |
+
"grad_norm": 4.95401899568707,
|
| 305 |
+
"learning_rate": 0.0001,
|
| 306 |
+
"loss": 0.051,
|
| 307 |
+
"num_input_tokens_seen": 1697280,
|
| 308 |
+
"step": 170
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.09013649240278135,
|
| 312 |
+
"grad_norm": 0.34558354886099124,
|
| 313 |
+
"learning_rate": 9.999940874631277e-05,
|
| 314 |
+
"loss": 0.0584,
|
| 315 |
+
"num_input_tokens_seen": 1747200,
|
| 316 |
+
"step": 175
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.09271182075714654,
|
| 320 |
+
"grad_norm": 3.6326401692458146,
|
| 321 |
+
"learning_rate": 9.999763499923432e-05,
|
| 322 |
+
"loss": 0.0704,
|
| 323 |
+
"num_input_tokens_seen": 1797120,
|
| 324 |
+
"step": 180
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.09528714911151172,
|
| 328 |
+
"grad_norm": 0.4999720825488852,
|
| 329 |
+
"learning_rate": 9.999467880071402e-05,
|
| 330 |
+
"loss": 0.0278,
|
| 331 |
+
"num_input_tokens_seen": 1847040,
|
| 332 |
+
"step": 185
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.0978624774658769,
|
| 336 |
+
"grad_norm": 0.5211964505880501,
|
| 337 |
+
"learning_rate": 9.999054022066641e-05,
|
| 338 |
+
"loss": 0.0862,
|
| 339 |
+
"num_input_tokens_seen": 1896960,
|
| 340 |
+
"step": 190
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.10043780582024209,
|
| 344 |
+
"grad_norm": 0.8767038751832389,
|
| 345 |
+
"learning_rate": 9.998521935696953e-05,
|
| 346 |
+
"loss": 0.0565,
|
| 347 |
+
"num_input_tokens_seen": 1946880,
|
| 348 |
+
"step": 195
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 0.10301313417460727,
|
| 352 |
+
"grad_norm": 0.1848827361202722,
|
| 353 |
+
"learning_rate": 9.997871633546257e-05,
|
| 354 |
+
"loss": 0.0459,
|
| 355 |
+
"num_input_tokens_seen": 1996800,
|
| 356 |
+
"step": 200
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.10301313417460727,
|
| 360 |
+
"eval_loss": 0.09957947582006454,
|
| 361 |
+
"eval_runtime": 19.2651,
|
| 362 |
+
"eval_samples_per_second": 3.114,
|
| 363 |
+
"eval_steps_per_second": 0.779,
|
| 364 |
+
"num_input_tokens_seen": 1996800,
|
| 365 |
+
"step": 200
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 0.10558846252897244,
|
| 369 |
+
"grad_norm": 1.0969393658164421,
|
| 370 |
+
"learning_rate": 9.997103130994296e-05,
|
| 371 |
+
"loss": 0.0539,
|
| 372 |
+
"num_input_tokens_seen": 2046720,
|
| 373 |
+
"step": 205
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.10816379088333762,
|
| 377 |
+
"grad_norm": 2.885869194934028,
|
| 378 |
+
"learning_rate": 9.996216446216267e-05,
|
| 379 |
+
"loss": 0.0654,
|
| 380 |
+
"num_input_tokens_seen": 2096640,
|
| 381 |
+
"step": 210
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.11073911923770281,
|
| 385 |
+
"grad_norm": 0.5225257245731217,
|
| 386 |
+
"learning_rate": 9.995211600182397e-05,
|
| 387 |
+
"loss": 0.0316,
|
| 388 |
+
"num_input_tokens_seen": 2146560,
|
| 389 |
+
"step": 215
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.11331444759206799,
|
| 393 |
+
"grad_norm": 2.1553510734212797,
|
| 394 |
+
"learning_rate": 9.994088616657444e-05,
|
| 395 |
+
"loss": 0.1169,
|
| 396 |
+
"num_input_tokens_seen": 2196480,
|
| 397 |
+
"step": 220
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 0.11588977594643317,
|
| 401 |
+
"grad_norm": 1.1133884703723633,
|
| 402 |
+
"learning_rate": 9.992847522200133e-05,
|
| 403 |
+
"loss": 0.0382,
|
| 404 |
+
"num_input_tokens_seen": 2246400,
|
| 405 |
+
"step": 225
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.11846510430079835,
|
| 409 |
+
"grad_norm": 0.8875243341616034,
|
| 410 |
+
"learning_rate": 9.99148834616253e-05,
|
| 411 |
+
"loss": 0.0406,
|
| 412 |
+
"num_input_tokens_seen": 2296320,
|
| 413 |
+
"step": 230
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.12104043265516354,
|
| 417 |
+
"grad_norm": 1.81283533812695,
|
| 418 |
+
"learning_rate": 9.990011120689351e-05,
|
| 419 |
+
"loss": 0.0182,
|
| 420 |
+
"num_input_tokens_seen": 2346240,
|
| 421 |
+
"step": 235
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.12361576100952872,
|
| 425 |
+
"grad_norm": 3.873083258671571,
|
| 426 |
+
"learning_rate": 9.988415880717194e-05,
|
| 427 |
+
"loss": 0.0881,
|
| 428 |
+
"num_input_tokens_seen": 2396160,
|
| 429 |
+
"step": 240
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.1261910893638939,
|
| 433 |
+
"grad_norm": 3.427761103620865,
|
| 434 |
+
"learning_rate": 9.986702663973722e-05,
|
| 435 |
+
"loss": 0.0565,
|
| 436 |
+
"num_input_tokens_seen": 2446080,
|
| 437 |
+
"step": 245
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.12876641771825909,
|
| 441 |
+
"grad_norm": 1.531943599765959,
|
| 442 |
+
"learning_rate": 9.98487151097676e-05,
|
| 443 |
+
"loss": 0.0805,
|
| 444 |
+
"num_input_tokens_seen": 2496000,
|
| 445 |
+
"step": 250
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.12876641771825909,
|
| 449 |
+
"eval_loss": 0.055875860154628754,
|
| 450 |
+
"eval_runtime": 19.5106,
|
| 451 |
+
"eval_samples_per_second": 3.075,
|
| 452 |
+
"eval_steps_per_second": 0.769,
|
| 453 |
+
"num_input_tokens_seen": 2496000,
|
| 454 |
+
"step": 250
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.13134174607262425,
|
| 458 |
+
"grad_norm": 0.845514751132877,
|
| 459 |
+
"learning_rate": 9.98292246503335e-05,
|
| 460 |
+
"loss": 0.0235,
|
| 461 |
+
"num_input_tokens_seen": 2545920,
|
| 462 |
+
"step": 255
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 0.13391707442698944,
|
| 466 |
+
"grad_norm": 1.7441538057875345,
|
| 467 |
+
"learning_rate": 9.980855572238714e-05,
|
| 468 |
+
"loss": 0.0137,
|
| 469 |
+
"num_input_tokens_seen": 2595840,
|
| 470 |
+
"step": 260
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.13649240278135463,
|
| 474 |
+
"grad_norm": 0.6628270146260998,
|
| 475 |
+
"learning_rate": 9.978670881475172e-05,
|
| 476 |
+
"loss": 0.0605,
|
| 477 |
+
"num_input_tokens_seen": 2645760,
|
| 478 |
+
"step": 265
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.1390677311357198,
|
| 482 |
+
"grad_norm": 4.49926360488578,
|
| 483 |
+
"learning_rate": 9.976368444410985e-05,
|
| 484 |
+
"loss": 0.0585,
|
| 485 |
+
"num_input_tokens_seen": 2695680,
|
| 486 |
+
"step": 270
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.141643059490085,
|
| 490 |
+
"grad_norm": 0.7159498856832776,
|
| 491 |
+
"learning_rate": 9.973948315499126e-05,
|
| 492 |
+
"loss": 0.0096,
|
| 493 |
+
"num_input_tokens_seen": 2745600,
|
| 494 |
+
"step": 275
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.14421838784445018,
|
| 498 |
+
"grad_norm": 4.09842502301765,
|
| 499 |
+
"learning_rate": 9.971410551976002e-05,
|
| 500 |
+
"loss": 0.1383,
|
| 501 |
+
"num_input_tokens_seen": 2795520,
|
| 502 |
+
"step": 280
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 0.14679371619881534,
|
| 506 |
+
"grad_norm": 1.0773897808674016,
|
| 507 |
+
"learning_rate": 9.968755213860094e-05,
|
| 508 |
+
"loss": 0.049,
|
| 509 |
+
"num_input_tokens_seen": 2845440,
|
| 510 |
+
"step": 285
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 0.14936904455318054,
|
| 514 |
+
"grad_norm": 1.7224091076306651,
|
| 515 |
+
"learning_rate": 9.96598236395054e-05,
|
| 516 |
+
"loss": 0.0336,
|
| 517 |
+
"num_input_tokens_seen": 2895360,
|
| 518 |
+
"step": 290
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.1519443729075457,
|
| 522 |
+
"grad_norm": 0.24158350585209873,
|
| 523 |
+
"learning_rate": 9.96309206782565e-05,
|
| 524 |
+
"loss": 0.019,
|
| 525 |
+
"num_input_tokens_seen": 2945280,
|
| 526 |
+
"step": 295
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.1545197012619109,
|
| 530 |
+
"grad_norm": 0.6614271496466305,
|
| 531 |
+
"learning_rate": 9.960084393841355e-05,
|
| 532 |
+
"loss": 0.0381,
|
| 533 |
+
"num_input_tokens_seen": 2995200,
|
| 534 |
+
"step": 300
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.1545197012619109,
|
| 538 |
+
"eval_loss": 0.030908752232789993,
|
| 539 |
+
"eval_runtime": 19.8702,
|
| 540 |
+
"eval_samples_per_second": 3.02,
|
| 541 |
+
"eval_steps_per_second": 0.755,
|
| 542 |
+
"num_input_tokens_seen": 2995200,
|
| 543 |
+
"step": 300
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 0.15709502961627608,
|
| 547 |
+
"grad_norm": 0.2314383428097956,
|
| 548 |
+
"learning_rate": 9.956959413129585e-05,
|
| 549 |
+
"loss": 0.0245,
|
| 550 |
+
"num_input_tokens_seen": 3045120,
|
| 551 |
+
"step": 305
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 0.15967035797064125,
|
| 555 |
+
"grad_norm": 0.2035043692790888,
|
| 556 |
+
"learning_rate": 9.953717199596598e-05,
|
| 557 |
+
"loss": 0.0121,
|
| 558 |
+
"num_input_tokens_seen": 3095040,
|
| 559 |
+
"step": 310
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 0.16224568632500644,
|
| 563 |
+
"grad_norm": 0.014152925203425593,
|
| 564 |
+
"learning_rate": 9.95035782992122e-05,
|
| 565 |
+
"loss": 0.054,
|
| 566 |
+
"num_input_tokens_seen": 3144960,
|
| 567 |
+
"step": 315
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 0.16482101467937163,
|
| 571 |
+
"grad_norm": 1.0162363271291994,
|
| 572 |
+
"learning_rate": 9.94688138355304e-05,
|
| 573 |
+
"loss": 0.0849,
|
| 574 |
+
"num_input_tokens_seen": 3194880,
|
| 575 |
+
"step": 320
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.1673963430337368,
|
| 579 |
+
"grad_norm": 3.92714662464831,
|
| 580 |
+
"learning_rate": 9.943287942710527e-05,
|
| 581 |
+
"loss": 0.0503,
|
| 582 |
+
"num_input_tokens_seen": 3244800,
|
| 583 |
+
"step": 325
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.16997167138810199,
|
| 587 |
+
"grad_norm": 0.47450325213422423,
|
| 588 |
+
"learning_rate": 9.939577592379088e-05,
|
| 589 |
+
"loss": 0.0302,
|
| 590 |
+
"num_input_tokens_seen": 3294720,
|
| 591 |
+
"step": 330
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.17254699974246718,
|
| 595 |
+
"grad_norm": 1.6706834486788253,
|
| 596 |
+
"learning_rate": 9.935750420309055e-05,
|
| 597 |
+
"loss": 0.082,
|
| 598 |
+
"num_input_tokens_seen": 3344640,
|
| 599 |
+
"step": 335
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 0.17512232809683234,
|
| 603 |
+
"grad_norm": 0.16856461555999938,
|
| 604 |
+
"learning_rate": 9.931806517013612e-05,
|
| 605 |
+
"loss": 0.0411,
|
| 606 |
+
"num_input_tokens_seen": 3394560,
|
| 607 |
+
"step": 340
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 0.17769765645119753,
|
| 611 |
+
"grad_norm": 8.475859938133665,
|
| 612 |
+
"learning_rate": 9.927745975766654e-05,
|
| 613 |
+
"loss": 0.1158,
|
| 614 |
+
"num_input_tokens_seen": 3444480,
|
| 615 |
+
"step": 345
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 0.1802729848055627,
|
| 619 |
+
"grad_norm": 0.7891476176494399,
|
| 620 |
+
"learning_rate": 9.923568892600578e-05,
|
| 621 |
+
"loss": 0.1761,
|
| 622 |
+
"num_input_tokens_seen": 3494400,
|
| 623 |
+
"step": 350
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 0.1802729848055627,
|
| 627 |
+
"eval_loss": 0.043948542326688766,
|
| 628 |
+
"eval_runtime": 19.3589,
|
| 629 |
+
"eval_samples_per_second": 3.099,
|
| 630 |
+
"eval_steps_per_second": 0.775,
|
| 631 |
+
"num_input_tokens_seen": 3494400,
|
| 632 |
+
"step": 350
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.1828483131599279,
|
| 636 |
+
"grad_norm": 0.3217035938609153,
|
| 637 |
+
"learning_rate": 9.91927536630402e-05,
|
| 638 |
+
"loss": 0.0267,
|
| 639 |
+
"num_input_tokens_seen": 3544320,
|
| 640 |
+
"step": 355
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.18542364151429308,
|
| 644 |
+
"grad_norm": 1.9146681345367045,
|
| 645 |
+
"learning_rate": 9.91486549841951e-05,
|
| 646 |
+
"loss": 0.0642,
|
| 647 |
+
"num_input_tokens_seen": 3594240,
|
| 648 |
+
"step": 360
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.18799896986865824,
|
| 652 |
+
"grad_norm": 0.053067921944857786,
|
| 653 |
+
"learning_rate": 9.91033939324107e-05,
|
| 654 |
+
"loss": 0.0302,
|
| 655 |
+
"num_input_tokens_seen": 3644160,
|
| 656 |
+
"step": 365
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.19057429822302344,
|
| 660 |
+
"grad_norm": 0.07277339857890938,
|
| 661 |
+
"learning_rate": 9.905697157811761e-05,
|
| 662 |
+
"loss": 0.0099,
|
| 663 |
+
"num_input_tokens_seen": 3694080,
|
| 664 |
+
"step": 370
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 0.19314962657738863,
|
| 668 |
+
"grad_norm": 4.117705571576552,
|
| 669 |
+
"learning_rate": 9.900938901921131e-05,
|
| 670 |
+
"loss": 0.0573,
|
| 671 |
+
"num_input_tokens_seen": 3744000,
|
| 672 |
+
"step": 375
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 0.1957249549317538,
|
| 676 |
+
"grad_norm": 2.4479374171708304,
|
| 677 |
+
"learning_rate": 9.896064738102635e-05,
|
| 678 |
+
"loss": 0.0302,
|
| 679 |
+
"num_input_tokens_seen": 3793920,
|
| 680 |
+
"step": 380
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.19830028328611898,
|
| 684 |
+
"grad_norm": 2.1751552571638153,
|
| 685 |
+
"learning_rate": 9.891074781630966e-05,
|
| 686 |
+
"loss": 0.0374,
|
| 687 |
+
"num_input_tokens_seen": 3843840,
|
| 688 |
+
"step": 385
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.20087561164048418,
|
| 692 |
+
"grad_norm": 1.4138134932494708,
|
| 693 |
+
"learning_rate": 9.885969150519331e-05,
|
| 694 |
+
"loss": 0.0227,
|
| 695 |
+
"num_input_tokens_seen": 3893760,
|
| 696 |
+
"step": 390
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.20345093999484934,
|
| 700 |
+
"grad_norm": 0.3312762825665537,
|
| 701 |
+
"learning_rate": 9.88074796551666e-05,
|
| 702 |
+
"loss": 0.0104,
|
| 703 |
+
"num_input_tokens_seen": 3943680,
|
| 704 |
+
"step": 395
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.20602626834921453,
|
| 708 |
+
"grad_norm": 0.06774783590394957,
|
| 709 |
+
"learning_rate": 9.875411350104744e-05,
|
| 710 |
+
"loss": 0.0146,
|
| 711 |
+
"num_input_tokens_seen": 3993600,
|
| 712 |
+
"step": 400
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 0.20602626834921453,
|
| 716 |
+
"eval_loss": 0.024433813989162445,
|
| 717 |
+
"eval_runtime": 18.8143,
|
| 718 |
+
"eval_samples_per_second": 3.189,
|
| 719 |
+
"eval_steps_per_second": 0.797,
|
| 720 |
+
"num_input_tokens_seen": 3993600,
|
| 721 |
+
"step": 400
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.2086015967035797,
|
| 725 |
+
"grad_norm": 0.06416132017781931,
|
| 726 |
+
"learning_rate": 9.86995943049533e-05,
|
| 727 |
+
"loss": 0.0074,
|
| 728 |
+
"num_input_tokens_seen": 4043520,
|
| 729 |
+
"step": 405
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.2111769250579449,
|
| 733 |
+
"grad_norm": 0.04809215529224866,
|
| 734 |
+
"learning_rate": 9.864392335627117e-05,
|
| 735 |
+
"loss": 0.0375,
|
| 736 |
+
"num_input_tokens_seen": 4093440,
|
| 737 |
+
"step": 410
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.21375225341231008,
|
| 741 |
+
"grad_norm": 0.40943321947836153,
|
| 742 |
+
"learning_rate": 9.858710197162721e-05,
|
| 743 |
+
"loss": 0.0217,
|
| 744 |
+
"num_input_tokens_seen": 4143360,
|
| 745 |
+
"step": 415
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.21632758176667524,
|
| 749 |
+
"grad_norm": 0.08069011480339518,
|
| 750 |
+
"learning_rate": 9.852913149485556e-05,
|
| 751 |
+
"loss": 0.016,
|
| 752 |
+
"num_input_tokens_seen": 4193280,
|
| 753 |
+
"step": 420
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.21890291012104043,
|
| 757 |
+
"grad_norm": 1.806585526467194,
|
| 758 |
+
"learning_rate": 9.847001329696653e-05,
|
| 759 |
+
"loss": 0.0832,
|
| 760 |
+
"num_input_tokens_seen": 4243200,
|
| 761 |
+
"step": 425
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.22147823847540563,
|
| 765 |
+
"grad_norm": 3.122914329597603,
|
| 766 |
+
"learning_rate": 9.840974877611422e-05,
|
| 767 |
+
"loss": 0.0444,
|
| 768 |
+
"num_input_tokens_seen": 4293120,
|
| 769 |
+
"step": 430
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 0.2240535668297708,
|
| 773 |
+
"grad_norm": 1.0101510644418257,
|
| 774 |
+
"learning_rate": 9.834833935756344e-05,
|
| 775 |
+
"loss": 0.0465,
|
| 776 |
+
"num_input_tokens_seen": 4343040,
|
| 777 |
+
"step": 435
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 0.22662889518413598,
|
| 781 |
+
"grad_norm": 1.7770164472545809,
|
| 782 |
+
"learning_rate": 9.828578649365601e-05,
|
| 783 |
+
"loss": 0.0428,
|
| 784 |
+
"num_input_tokens_seen": 4392960,
|
| 785 |
+
"step": 440
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.22920422353850115,
|
| 789 |
+
"grad_norm": 0.9798524063329249,
|
| 790 |
+
"learning_rate": 9.822209166377635e-05,
|
| 791 |
+
"loss": 0.02,
|
| 792 |
+
"num_input_tokens_seen": 4442880,
|
| 793 |
+
"step": 445
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.23177955189286634,
|
| 797 |
+
"grad_norm": 0.22305429096692395,
|
| 798 |
+
"learning_rate": 9.815725637431662e-05,
|
| 799 |
+
"loss": 0.0157,
|
| 800 |
+
"num_input_tokens_seen": 4492800,
|
| 801 |
+
"step": 450
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.23177955189286634,
|
| 805 |
+
"eval_loss": 0.006690301466733217,
|
| 806 |
+
"eval_runtime": 18.7725,
|
| 807 |
+
"eval_samples_per_second": 3.196,
|
| 808 |
+
"eval_steps_per_second": 0.799,
|
| 809 |
+
"num_input_tokens_seen": 4492800,
|
| 810 |
+
"step": 450
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 0.23435488024723153,
|
| 814 |
+
"grad_norm": 2.8137229294991064,
|
| 815 |
+
"learning_rate": 9.809128215864097e-05,
|
| 816 |
+
"loss": 0.0596,
|
| 817 |
+
"num_input_tokens_seen": 4542720,
|
| 818 |
+
"step": 455
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 0.2369302086015967,
|
| 822 |
+
"grad_norm": 0.045182426772616975,
|
| 823 |
+
"learning_rate": 9.802417057704931e-05,
|
| 824 |
+
"loss": 0.0238,
|
| 825 |
+
"num_input_tokens_seen": 4592640,
|
| 826 |
+
"step": 460
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 0.23950553695596188,
|
| 830 |
+
"grad_norm": 2.1659990503080424,
|
| 831 |
+
"learning_rate": 9.795592321674045e-05,
|
| 832 |
+
"loss": 0.0331,
|
| 833 |
+
"num_input_tokens_seen": 4642560,
|
| 834 |
+
"step": 465
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.24208086531032708,
|
| 838 |
+
"grad_norm": 0.3208179936186945,
|
| 839 |
+
"learning_rate": 9.788654169177453e-05,
|
| 840 |
+
"loss": 0.0177,
|
| 841 |
+
"num_input_tokens_seen": 4692480,
|
| 842 |
+
"step": 470
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.24465619366469224,
|
| 846 |
+
"grad_norm": 1.4150868022870595,
|
| 847 |
+
"learning_rate": 9.781602764303487e-05,
|
| 848 |
+
"loss": 0.028,
|
| 849 |
+
"num_input_tokens_seen": 4742400,
|
| 850 |
+
"step": 475
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.24723152201905743,
|
| 854 |
+
"grad_norm": 0.21015675310797993,
|
| 855 |
+
"learning_rate": 9.774438273818911e-05,
|
| 856 |
+
"loss": 0.0561,
|
| 857 |
+
"num_input_tokens_seen": 4792320,
|
| 858 |
+
"step": 480
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 0.24980685037342262,
|
| 862 |
+
"grad_norm": 2.6788940269709287,
|
| 863 |
+
"learning_rate": 9.767160867164979e-05,
|
| 864 |
+
"loss": 0.0242,
|
| 865 |
+
"num_input_tokens_seen": 4842240,
|
| 866 |
+
"step": 485
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.2523821787277878,
|
| 870 |
+
"grad_norm": 0.03933517764434036,
|
| 871 |
+
"learning_rate": 9.759770716453436e-05,
|
| 872 |
+
"loss": 0.0286,
|
| 873 |
+
"num_input_tokens_seen": 4892160,
|
| 874 |
+
"step": 490
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 0.254957507082153,
|
| 878 |
+
"grad_norm": 2.640299598300008,
|
| 879 |
+
"learning_rate": 9.752267996462434e-05,
|
| 880 |
+
"loss": 0.0394,
|
| 881 |
+
"num_input_tokens_seen": 4942080,
|
| 882 |
+
"step": 495
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 0.25753283543651817,
|
| 886 |
+
"grad_norm": 0.03143215150024672,
|
| 887 |
+
"learning_rate": 9.744652884632406e-05,
|
| 888 |
+
"loss": 0.0122,
|
| 889 |
+
"num_input_tokens_seen": 4992000,
|
| 890 |
+
"step": 500
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.25753283543651817,
|
| 894 |
+
"eval_loss": 0.007972972467541695,
|
| 895 |
+
"eval_runtime": 18.8509,
|
| 896 |
+
"eval_samples_per_second": 3.183,
|
| 897 |
+
"eval_steps_per_second": 0.796,
|
| 898 |
+
"num_input_tokens_seen": 4992000,
|
| 899 |
+
"step": 500
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.26010816379088336,
|
| 903 |
+
"grad_norm": 0.3421379771468272,
|
| 904 |
+
"learning_rate": 9.736925561061871e-05,
|
| 905 |
+
"loss": 0.0184,
|
| 906 |
+
"num_input_tokens_seen": 5041920,
|
| 907 |
+
"step": 505
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 0.2626834921452485,
|
| 911 |
+
"grad_norm": 0.16336037200955245,
|
| 912 |
+
"learning_rate": 9.729086208503174e-05,
|
| 913 |
+
"loss": 0.0046,
|
| 914 |
+
"num_input_tokens_seen": 5091840,
|
| 915 |
+
"step": 510
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 0.2652588204996137,
|
| 919 |
+
"grad_norm": 0.7318312729236729,
|
| 920 |
+
"learning_rate": 9.721135012358156e-05,
|
| 921 |
+
"loss": 0.0226,
|
| 922 |
+
"num_input_tokens_seen": 5141760,
|
| 923 |
+
"step": 515
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 0.2678341488539789,
|
| 927 |
+
"grad_norm": 2.1703341770378333,
|
| 928 |
+
"learning_rate": 9.713072160673777e-05,
|
| 929 |
+
"loss": 0.0475,
|
| 930 |
+
"num_input_tokens_seen": 5191680,
|
| 931 |
+
"step": 520
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 0.2704094772083441,
|
| 935 |
+
"grad_norm": 1.6372493711976897,
|
| 936 |
+
"learning_rate": 9.704897844137673e-05,
|
| 937 |
+
"loss": 0.0572,
|
| 938 |
+
"num_input_tokens_seen": 5241600,
|
| 939 |
+
"step": 525
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.27298480556270927,
|
| 943 |
+
"grad_norm": 3.5840623375725804,
|
| 944 |
+
"learning_rate": 9.696612256073633e-05,
|
| 945 |
+
"loss": 0.025,
|
| 946 |
+
"num_input_tokens_seen": 5291520,
|
| 947 |
+
"step": 530
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.2755601339170744,
|
| 951 |
+
"grad_norm": 0.06141348661076509,
|
| 952 |
+
"learning_rate": 9.688215592437039e-05,
|
| 953 |
+
"loss": 0.0122,
|
| 954 |
+
"num_input_tokens_seen": 5341440,
|
| 955 |
+
"step": 535
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.2781354622714396,
|
| 959 |
+
"grad_norm": 0.04980492840656424,
|
| 960 |
+
"learning_rate": 9.679708051810221e-05,
|
| 961 |
+
"loss": 0.0045,
|
| 962 |
+
"num_input_tokens_seen": 5391360,
|
| 963 |
+
"step": 540
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 0.2807107906258048,
|
| 967 |
+
"grad_norm": 3.153079438616822,
|
| 968 |
+
"learning_rate": 9.67108983539777e-05,
|
| 969 |
+
"loss": 0.0431,
|
| 970 |
+
"num_input_tokens_seen": 5441280,
|
| 971 |
+
"step": 545
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 0.28328611898017,
|
| 975 |
+
"grad_norm": 1.6764625232464418,
|
| 976 |
+
"learning_rate": 9.662361147021779e-05,
|
| 977 |
+
"loss": 0.0339,
|
| 978 |
+
"num_input_tokens_seen": 5491200,
|
| 979 |
+
"step": 550
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 0.28328611898017,
|
| 983 |
+
"eval_loss": 0.0033526704646646976,
|
| 984 |
+
"eval_runtime": 18.8387,
|
| 985 |
+
"eval_samples_per_second": 3.185,
|
| 986 |
+
"eval_steps_per_second": 0.796,
|
| 987 |
+
"num_input_tokens_seen": 5491200,
|
| 988 |
+
"step": 550
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 0.28586144733453517,
|
| 992 |
+
"grad_norm": 0.05427488923085365,
|
| 993 |
+
"learning_rate": 9.653522193117013e-05,
|
| 994 |
+
"loss": 0.0341,
|
| 995 |
+
"num_input_tokens_seen": 5541120,
|
| 996 |
+
"step": 555
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.28843677568890036,
|
| 1000 |
+
"grad_norm": 1.538376531568411,
|
| 1001 |
+
"learning_rate": 9.644573182726035e-05,
|
| 1002 |
+
"loss": 0.0341,
|
| 1003 |
+
"num_input_tokens_seen": 5591040,
|
| 1004 |
+
"step": 560
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.2910121040432655,
|
| 1008 |
+
"grad_norm": 0.04402865983757039,
|
| 1009 |
+
"learning_rate": 9.63551432749426e-05,
|
| 1010 |
+
"loss": 0.0453,
|
| 1011 |
+
"num_input_tokens_seen": 5640960,
|
| 1012 |
+
"step": 565
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 0.2935874323976307,
|
| 1016 |
+
"grad_norm": 0.1469885874710427,
|
| 1017 |
+
"learning_rate": 9.626345841664953e-05,
|
| 1018 |
+
"loss": 0.0227,
|
| 1019 |
+
"num_input_tokens_seen": 5690880,
|
| 1020 |
+
"step": 570
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 0.2961627607519959,
|
| 1024 |
+
"grad_norm": 0.3459204695983954,
|
| 1025 |
+
"learning_rate": 9.617067942074153e-05,
|
| 1026 |
+
"loss": 0.0135,
|
| 1027 |
+
"num_input_tokens_seen": 5740800,
|
| 1028 |
+
"step": 575
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 0.29873808910636107,
|
| 1032 |
+
"grad_norm": 1.7604264611882452,
|
| 1033 |
+
"learning_rate": 9.607680848145558e-05,
|
| 1034 |
+
"loss": 0.0439,
|
| 1035 |
+
"num_input_tokens_seen": 5790720,
|
| 1036 |
+
"step": 580
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 0.30131341746072626,
|
| 1040 |
+
"grad_norm": 2.1661707703583204,
|
| 1041 |
+
"learning_rate": 9.598184781885318e-05,
|
| 1042 |
+
"loss": 0.0344,
|
| 1043 |
+
"num_input_tokens_seen": 5840640,
|
| 1044 |
+
"step": 585
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.3038887458150914,
|
| 1048 |
+
"grad_norm": 0.19674934066326408,
|
| 1049 |
+
"learning_rate": 9.588579967876806e-05,
|
| 1050 |
+
"loss": 0.0244,
|
| 1051 |
+
"num_input_tokens_seen": 5890560,
|
| 1052 |
+
"step": 590
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.3064640741694566,
|
| 1056 |
+
"grad_norm": 4.463551986915384,
|
| 1057 |
+
"learning_rate": 9.578866633275288e-05,
|
| 1058 |
+
"loss": 0.022,
|
| 1059 |
+
"num_input_tokens_seen": 5940480,
|
| 1060 |
+
"step": 595
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.3090394025238218,
|
| 1064 |
+
"grad_norm": 0.08358626804613765,
|
| 1065 |
+
"learning_rate": 9.569045007802559e-05,
|
| 1066 |
+
"loss": 0.0217,
|
| 1067 |
+
"num_input_tokens_seen": 5990400,
|
| 1068 |
+
"step": 600
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 0.3090394025238218,
|
| 1072 |
+
"eval_loss": 0.013290103524923325,
|
| 1073 |
+
"eval_runtime": 18.8576,
|
| 1074 |
+
"eval_samples_per_second": 3.182,
|
| 1075 |
+
"eval_steps_per_second": 0.795,
|
| 1076 |
+
"num_input_tokens_seen": 5990400,
|
| 1077 |
+
"step": 600
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 0.311614730878187,
|
| 1081 |
+
"grad_norm": 0.06328082378687645,
|
| 1082 |
+
"learning_rate": 9.55911532374151e-05,
|
| 1083 |
+
"loss": 0.0207,
|
| 1084 |
+
"num_input_tokens_seen": 6040320,
|
| 1085 |
+
"step": 605
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 0.31419005923255217,
|
| 1089 |
+
"grad_norm": 1.846961607001629,
|
| 1090 |
+
"learning_rate": 9.549077815930636e-05,
|
| 1091 |
+
"loss": 0.023,
|
| 1092 |
+
"num_input_tokens_seen": 6090240,
|
| 1093 |
+
"step": 610
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 0.31676538758691736,
|
| 1097 |
+
"grad_norm": 0.07122251919235507,
|
| 1098 |
+
"learning_rate": 9.538932721758474e-05,
|
| 1099 |
+
"loss": 0.017,
|
| 1100 |
+
"num_input_tokens_seen": 6140160,
|
| 1101 |
+
"step": 615
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.3193407159412825,
|
| 1105 |
+
"grad_norm": 1.9564639611342651,
|
| 1106 |
+
"learning_rate": 9.528680281157999e-05,
|
| 1107 |
+
"loss": 0.009,
|
| 1108 |
+
"num_input_tokens_seen": 6190080,
|
| 1109 |
+
"step": 620
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.3219160442956477,
|
| 1113 |
+
"grad_norm": 0.43628623054662674,
|
| 1114 |
+
"learning_rate": 9.518320736600943e-05,
|
| 1115 |
+
"loss": 0.0421,
|
| 1116 |
+
"num_input_tokens_seen": 6240000,
|
| 1117 |
+
"step": 625
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 0.3244913726500129,
|
| 1121 |
+
"grad_norm": 0.7626652394047067,
|
| 1122 |
+
"learning_rate": 9.507854333092063e-05,
|
| 1123 |
+
"loss": 0.0324,
|
| 1124 |
+
"num_input_tokens_seen": 6289920,
|
| 1125 |
+
"step": 630
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 0.32706670100437807,
|
| 1129 |
+
"grad_norm": 0.5823990007842583,
|
| 1130 |
+
"learning_rate": 9.497281318163346e-05,
|
| 1131 |
+
"loss": 0.0139,
|
| 1132 |
+
"num_input_tokens_seen": 6339840,
|
| 1133 |
+
"step": 635
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 0.32964202935874326,
|
| 1137 |
+
"grad_norm": 1.6726620150276597,
|
| 1138 |
+
"learning_rate": 9.486601941868154e-05,
|
| 1139 |
+
"loss": 0.0375,
|
| 1140 |
+
"num_input_tokens_seen": 6389760,
|
| 1141 |
+
"step": 640
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 0.3322173577131084,
|
| 1145 |
+
"grad_norm": 0.08456122780891835,
|
| 1146 |
+
"learning_rate": 9.475816456775313e-05,
|
| 1147 |
+
"loss": 0.0618,
|
| 1148 |
+
"num_input_tokens_seen": 6439680,
|
| 1149 |
+
"step": 645
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 0.3347926860674736,
|
| 1153 |
+
"grad_norm": 0.3287256358007373,
|
| 1154 |
+
"learning_rate": 9.464925117963133e-05,
|
| 1155 |
+
"loss": 0.0327,
|
| 1156 |
+
"num_input_tokens_seen": 6489600,
|
| 1157 |
+
"step": 650
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.3347926860674736,
|
| 1161 |
+
"eval_loss": 0.02095886506140232,
|
| 1162 |
+
"eval_runtime": 19.1993,
|
| 1163 |
+
"eval_samples_per_second": 3.125,
|
| 1164 |
+
"eval_steps_per_second": 0.781,
|
| 1165 |
+
"num_input_tokens_seen": 6489600,
|
| 1166 |
+
"step": 650
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 0.3373680144218388,
|
| 1170 |
+
"grad_norm": 1.642912250249591,
|
| 1171 |
+
"learning_rate": 9.453928183013385e-05,
|
| 1172 |
+
"loss": 0.037,
|
| 1173 |
+
"num_input_tokens_seen": 6539520,
|
| 1174 |
+
"step": 655
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 0.33994334277620397,
|
| 1178 |
+
"grad_norm": 1.6829194140719153,
|
| 1179 |
+
"learning_rate": 9.442825912005202e-05,
|
| 1180 |
+
"loss": 0.034,
|
| 1181 |
+
"num_input_tokens_seen": 6589440,
|
| 1182 |
+
"step": 660
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 0.34251867113056916,
|
| 1186 |
+
"grad_norm": 1.7722553129499397,
|
| 1187 |
+
"learning_rate": 9.431618567508933e-05,
|
| 1188 |
+
"loss": 0.0584,
|
| 1189 |
+
"num_input_tokens_seen": 6639360,
|
| 1190 |
+
"step": 665
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 0.34509399948493436,
|
| 1194 |
+
"grad_norm": 1.8398383207694327,
|
| 1195 |
+
"learning_rate": 9.420306414579925e-05,
|
| 1196 |
+
"loss": 0.059,
|
| 1197 |
+
"num_input_tokens_seen": 6689280,
|
| 1198 |
+
"step": 670
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 0.3476693278392995,
|
| 1202 |
+
"grad_norm": 0.2451231053960763,
|
| 1203 |
+
"learning_rate": 9.408889720752266e-05,
|
| 1204 |
+
"loss": 0.023,
|
| 1205 |
+
"num_input_tokens_seen": 6739200,
|
| 1206 |
+
"step": 675
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.3502446561936647,
|
| 1210 |
+
"grad_norm": 0.13166014546250449,
|
| 1211 |
+
"learning_rate": 9.397368756032445e-05,
|
| 1212 |
+
"loss": 0.0327,
|
| 1213 |
+
"num_input_tokens_seen": 6789120,
|
| 1214 |
+
"step": 680
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.3528199845480299,
|
| 1218 |
+
"grad_norm": 1.9392864055798256,
|
| 1219 |
+
"learning_rate": 9.385743792892982e-05,
|
| 1220 |
+
"loss": 0.0241,
|
| 1221 |
+
"num_input_tokens_seen": 6839040,
|
| 1222 |
+
"step": 685
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 0.35539531290239507,
|
| 1226 |
+
"grad_norm": 0.7574757741852987,
|
| 1227 |
+
"learning_rate": 9.374015106265968e-05,
|
| 1228 |
+
"loss": 0.0227,
|
| 1229 |
+
"num_input_tokens_seen": 6888960,
|
| 1230 |
+
"step": 690
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 0.35797064125676026,
|
| 1234 |
+
"grad_norm": 1.1649279233149774,
|
| 1235 |
+
"learning_rate": 9.362182973536569e-05,
|
| 1236 |
+
"loss": 0.0171,
|
| 1237 |
+
"num_input_tokens_seen": 6938880,
|
| 1238 |
+
"step": 695
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 0.3605459696111254,
|
| 1242 |
+
"grad_norm": 0.04561814221227265,
|
| 1243 |
+
"learning_rate": 9.35024767453647e-05,
|
| 1244 |
+
"loss": 0.0267,
|
| 1245 |
+
"num_input_tokens_seen": 6988800,
|
| 1246 |
+
"step": 700
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 0.3605459696111254,
|
| 1250 |
+
"eval_loss": 0.005286735948175192,
|
| 1251 |
+
"eval_runtime": 19.2824,
|
| 1252 |
+
"eval_samples_per_second": 3.112,
|
| 1253 |
+
"eval_steps_per_second": 0.778,
|
| 1254 |
+
"num_input_tokens_seen": 6988800,
|
| 1255 |
+
"step": 700
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.3631212979654906,
|
| 1259 |
+
"grad_norm": 1.7432145241229078,
|
| 1260 |
+
"learning_rate": 9.338209491537257e-05,
|
| 1261 |
+
"loss": 0.0335,
|
| 1262 |
+
"num_input_tokens_seen": 7038720,
|
| 1263 |
+
"step": 705
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.3656966263198558,
|
| 1267 |
+
"grad_norm": 1.821635752664896,
|
| 1268 |
+
"learning_rate": 9.326068709243727e-05,
|
| 1269 |
+
"loss": 0.0108,
|
| 1270 |
+
"num_input_tokens_seen": 7088640,
|
| 1271 |
+
"step": 710
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 0.36827195467422097,
|
| 1275 |
+
"grad_norm": 0.5782878283043246,
|
| 1276 |
+
"learning_rate": 9.313825614787177e-05,
|
| 1277 |
+
"loss": 0.0296,
|
| 1278 |
+
"num_input_tokens_seen": 7138560,
|
| 1279 |
+
"step": 715
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 0.37084728302858616,
|
| 1283 |
+
"grad_norm": 2.0568373189070908,
|
| 1284 |
+
"learning_rate": 9.301480497718593e-05,
|
| 1285 |
+
"loss": 0.042,
|
| 1286 |
+
"num_input_tokens_seen": 7188480,
|
| 1287 |
+
"step": 720
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 0.37342261138295135,
|
| 1291 |
+
"grad_norm": 0.20623581315280795,
|
| 1292 |
+
"learning_rate": 9.289033650001817e-05,
|
| 1293 |
+
"loss": 0.0083,
|
| 1294 |
+
"num_input_tokens_seen": 7238400,
|
| 1295 |
+
"step": 725
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 0.3759979397373165,
|
| 1299 |
+
"grad_norm": 1.424505040979842,
|
| 1300 |
+
"learning_rate": 9.276485366006634e-05,
|
| 1301 |
+
"loss": 0.0109,
|
| 1302 |
+
"num_input_tokens_seen": 7288320,
|
| 1303 |
+
"step": 730
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 0.3785732680916817,
|
| 1307 |
+
"grad_norm": 2.1248350485820584,
|
| 1308 |
+
"learning_rate": 9.263835942501807e-05,
|
| 1309 |
+
"loss": 0.0122,
|
| 1310 |
+
"num_input_tokens_seen": 7338240,
|
| 1311 |
+
"step": 735
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.3811485964460469,
|
| 1315 |
+
"grad_norm": 0.357888255955218,
|
| 1316 |
+
"learning_rate": 9.251085678648072e-05,
|
| 1317 |
+
"loss": 0.017,
|
| 1318 |
+
"num_input_tokens_seen": 7388160,
|
| 1319 |
+
"step": 740
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.38372392480041206,
|
| 1323 |
+
"grad_norm": 3.859418562917624,
|
| 1324 |
+
"learning_rate": 9.238234875991046e-05,
|
| 1325 |
+
"loss": 0.0236,
|
| 1326 |
+
"num_input_tokens_seen": 7438080,
|
| 1327 |
+
"step": 745
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 0.38629925315477726,
|
| 1331 |
+
"grad_norm": 2.660896728773248,
|
| 1332 |
+
"learning_rate": 9.225283838454111e-05,
|
| 1333 |
+
"loss": 0.014,
|
| 1334 |
+
"num_input_tokens_seen": 7488000,
|
| 1335 |
+
"step": 750
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 0.38629925315477726,
|
| 1339 |
+
"eval_loss": 0.005334165878593922,
|
| 1340 |
+
"eval_runtime": 19.3986,
|
| 1341 |
+
"eval_samples_per_second": 3.093,
|
| 1342 |
+
"eval_steps_per_second": 0.773,
|
| 1343 |
+
"num_input_tokens_seen": 7488000,
|
| 1344 |
+
"step": 750
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 0.3888745815091424,
|
| 1348 |
+
"grad_norm": 3.511127189174474,
|
| 1349 |
+
"learning_rate": 9.21223287233121e-05,
|
| 1350 |
+
"loss": 0.0426,
|
| 1351 |
+
"num_input_tokens_seen": 7537920,
|
| 1352 |
+
"step": 755
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 0.3914499098635076,
|
| 1356 |
+
"grad_norm": 2.858952603531337,
|
| 1357 |
+
"learning_rate": 9.199082286279622e-05,
|
| 1358 |
+
"loss": 0.0094,
|
| 1359 |
+
"num_input_tokens_seen": 7587840,
|
| 1360 |
+
"step": 760
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.3940252382178728,
|
| 1364 |
+
"grad_norm": 1.9568584051828664,
|
| 1365 |
+
"learning_rate": 9.185832391312644e-05,
|
| 1366 |
+
"loss": 0.008,
|
| 1367 |
+
"num_input_tokens_seen": 7637760,
|
| 1368 |
+
"step": 765
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.39660056657223797,
|
| 1372 |
+
"grad_norm": 1.7485145115139782,
|
| 1373 |
+
"learning_rate": 9.172483500792244e-05,
|
| 1374 |
+
"loss": 0.0297,
|
| 1375 |
+
"num_input_tokens_seen": 7687680,
|
| 1376 |
+
"step": 770
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 0.39917589492660316,
|
| 1380 |
+
"grad_norm": 0.05161945290023201,
|
| 1381 |
+
"learning_rate": 9.159035930421658e-05,
|
| 1382 |
+
"loss": 0.0239,
|
| 1383 |
+
"num_input_tokens_seen": 7737600,
|
| 1384 |
+
"step": 775
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 0.40175122328096835,
|
| 1388 |
+
"grad_norm": 5.343080934932771,
|
| 1389 |
+
"learning_rate": 9.145489998237902e-05,
|
| 1390 |
+
"loss": 0.0162,
|
| 1391 |
+
"num_input_tokens_seen": 7787520,
|
| 1392 |
+
"step": 780
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 0.4043265516353335,
|
| 1396 |
+
"grad_norm": 0.020924688334471742,
|
| 1397 |
+
"learning_rate": 9.131846024604274e-05,
|
| 1398 |
+
"loss": 0.0309,
|
| 1399 |
+
"num_input_tokens_seen": 7837440,
|
| 1400 |
+
"step": 785
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 0.4069018799896987,
|
| 1404 |
+
"grad_norm": 0.5607043400469908,
|
| 1405 |
+
"learning_rate": 9.11810433220276e-05,
|
| 1406 |
+
"loss": 0.0432,
|
| 1407 |
+
"num_input_tokens_seen": 7887360,
|
| 1408 |
+
"step": 790
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 0.40947720834406387,
|
| 1412 |
+
"grad_norm": 0.05221845282486231,
|
| 1413 |
+
"learning_rate": 9.104265246026415e-05,
|
| 1414 |
+
"loss": 0.0022,
|
| 1415 |
+
"num_input_tokens_seen": 7937280,
|
| 1416 |
+
"step": 795
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.41205253669842906,
|
| 1420 |
+
"grad_norm": 0.09825330413354653,
|
| 1421 |
+
"learning_rate": 9.090329093371666e-05,
|
| 1422 |
+
"loss": 0.0065,
|
| 1423 |
+
"num_input_tokens_seen": 7987200,
|
| 1424 |
+
"step": 800
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.41205253669842906,
|
| 1428 |
+
"eval_loss": 0.006797688081860542,
|
| 1429 |
+
"eval_runtime": 19.3386,
|
| 1430 |
+
"eval_samples_per_second": 3.103,
|
| 1431 |
+
"eval_steps_per_second": 0.776,
|
| 1432 |
+
"num_input_tokens_seen": 7987200,
|
| 1433 |
+
"step": 800
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 0.41462786505279425,
|
| 1437 |
+
"grad_norm": 0.016984614729716928,
|
| 1438 |
+
"learning_rate": 9.076296203830579e-05,
|
| 1439 |
+
"loss": 0.0025,
|
| 1440 |
+
"num_input_tokens_seen": 8037120,
|
| 1441 |
+
"step": 805
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 0.4172031934071594,
|
| 1445 |
+
"grad_norm": 8.87156169406028,
|
| 1446 |
+
"learning_rate": 9.062166909283062e-05,
|
| 1447 |
+
"loss": 0.0338,
|
| 1448 |
+
"num_input_tokens_seen": 8087040,
|
| 1449 |
+
"step": 810
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 0.4197785217615246,
|
| 1453 |
+
"grad_norm": 0.07123492673307173,
|
| 1454 |
+
"learning_rate": 9.047941543889014e-05,
|
| 1455 |
+
"loss": 0.0505,
|
| 1456 |
+
"num_input_tokens_seen": 8136960,
|
| 1457 |
+
"step": 815
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 0.4223538501158898,
|
| 1461 |
+
"grad_norm": 0.3486611879887143,
|
| 1462 |
+
"learning_rate": 9.033620444080428e-05,
|
| 1463 |
+
"loss": 0.0335,
|
| 1464 |
+
"num_input_tokens_seen": 8186880,
|
| 1465 |
+
"step": 820
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.42492917847025496,
|
| 1469 |
+
"grad_norm": 0.2658900406741178,
|
| 1470 |
+
"learning_rate": 9.019203948553422e-05,
|
| 1471 |
+
"loss": 0.0196,
|
| 1472 |
+
"num_input_tokens_seen": 8236800,
|
| 1473 |
+
"step": 825
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.42750450682462016,
|
| 1477 |
+
"grad_norm": 0.15327530188348007,
|
| 1478 |
+
"learning_rate": 9.004692398260244e-05,
|
| 1479 |
+
"loss": 0.018,
|
| 1480 |
+
"num_input_tokens_seen": 8286720,
|
| 1481 |
+
"step": 830
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 0.43007983517898535,
|
| 1485 |
+
"grad_norm": 5.292901102191953,
|
| 1486 |
+
"learning_rate": 8.9900861364012e-05,
|
| 1487 |
+
"loss": 0.012,
|
| 1488 |
+
"num_input_tokens_seen": 8336640,
|
| 1489 |
+
"step": 835
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 0.4326551635333505,
|
| 1493 |
+
"grad_norm": 3.2113529029102375,
|
| 1494 |
+
"learning_rate": 8.975385508416532e-05,
|
| 1495 |
+
"loss": 0.0062,
|
| 1496 |
+
"num_input_tokens_seen": 8386560,
|
| 1497 |
+
"step": 840
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 0.4352304918877157,
|
| 1501 |
+
"grad_norm": 0.3685880212421972,
|
| 1502 |
+
"learning_rate": 8.960590861978265e-05,
|
| 1503 |
+
"loss": 0.0232,
|
| 1504 |
+
"num_input_tokens_seen": 8436480,
|
| 1505 |
+
"step": 845
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 0.43780582024208087,
|
| 1509 |
+
"grad_norm": 1.4858234531184304,
|
| 1510 |
+
"learning_rate": 8.945702546981969e-05,
|
| 1511 |
+
"loss": 0.0306,
|
| 1512 |
+
"num_input_tokens_seen": 8486400,
|
| 1513 |
+
"step": 850
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 0.43780582024208087,
|
| 1517 |
+
"eval_loss": 0.007198736071586609,
|
| 1518 |
+
"eval_runtime": 19.7019,
|
| 1519 |
+
"eval_samples_per_second": 3.045,
|
| 1520 |
+
"eval_steps_per_second": 0.761,
|
| 1521 |
+
"num_input_tokens_seen": 8486400,
|
| 1522 |
+
"step": 850
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.44038114859644606,
|
| 1526 |
+
"grad_norm": 0.22602261311014887,
|
| 1527 |
+
"learning_rate": 8.930720915538487e-05,
|
| 1528 |
+
"loss": 0.0255,
|
| 1529 |
+
"num_input_tokens_seen": 8536320,
|
| 1530 |
+
"step": 855
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 0.44295647695081125,
|
| 1534 |
+
"grad_norm": 0.1515497727795517,
|
| 1535 |
+
"learning_rate": 8.915646321965614e-05,
|
| 1536 |
+
"loss": 0.0267,
|
| 1537 |
+
"num_input_tokens_seen": 8586240,
|
| 1538 |
+
"step": 860
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 0.4455318053051764,
|
| 1542 |
+
"grad_norm": 0.039053785243136956,
|
| 1543 |
+
"learning_rate": 8.900479122779712e-05,
|
| 1544 |
+
"loss": 0.0107,
|
| 1545 |
+
"num_input_tokens_seen": 8636160,
|
| 1546 |
+
"step": 865
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 0.4481071336595416,
|
| 1550 |
+
"grad_norm": 0.42882717357803835,
|
| 1551 |
+
"learning_rate": 8.885219676687277e-05,
|
| 1552 |
+
"loss": 0.0277,
|
| 1553 |
+
"num_input_tokens_seen": 8686080,
|
| 1554 |
+
"step": 870
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 0.45068246201390677,
|
| 1558 |
+
"grad_norm": 5.519940082555776,
|
| 1559 |
+
"learning_rate": 8.869868344576459e-05,
|
| 1560 |
+
"loss": 0.0175,
|
| 1561 |
+
"num_input_tokens_seen": 8736000,
|
| 1562 |
+
"step": 875
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 0.45325779036827196,
|
| 1566 |
+
"grad_norm": 0.15983443395533375,
|
| 1567 |
+
"learning_rate": 8.854425489508532e-05,
|
| 1568 |
+
"loss": 0.0065,
|
| 1569 |
+
"num_input_tokens_seen": 8785920,
|
| 1570 |
+
"step": 880
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.45583311872263715,
|
| 1574 |
+
"grad_norm": 0.42885948370195226,
|
| 1575 |
+
"learning_rate": 8.838891476709288e-05,
|
| 1576 |
+
"loss": 0.0476,
|
| 1577 |
+
"num_input_tokens_seen": 8835840,
|
| 1578 |
+
"step": 885
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.4584084470770023,
|
| 1582 |
+
"grad_norm": 0.18209352469474613,
|
| 1583 |
+
"learning_rate": 8.823266673560426e-05,
|
| 1584 |
+
"loss": 0.018,
|
| 1585 |
+
"num_input_tokens_seen": 8885760,
|
| 1586 |
+
"step": 890
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 0.4609837754313675,
|
| 1590 |
+
"grad_norm": 1.7895291298260116,
|
| 1591 |
+
"learning_rate": 8.807551449590846e-05,
|
| 1592 |
+
"loss": 0.0337,
|
| 1593 |
+
"num_input_tokens_seen": 8935680,
|
| 1594 |
+
"step": 895
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 0.4635591037857327,
|
| 1598 |
+
"grad_norm": 0.11579050055375595,
|
| 1599 |
+
"learning_rate": 8.791746176467907e-05,
|
| 1600 |
+
"loss": 0.0063,
|
| 1601 |
+
"num_input_tokens_seen": 8985600,
|
| 1602 |
+
"step": 900
|
| 1603 |
+
},
|
| 1604 |
+
{
|
| 1605 |
+
"epoch": 0.4635591037857327,
|
| 1606 |
+
"eval_loss": 0.010749292559921741,
|
| 1607 |
+
"eval_runtime": 19.2918,
|
| 1608 |
+
"eval_samples_per_second": 3.11,
|
| 1609 |
+
"eval_steps_per_second": 0.778,
|
| 1610 |
+
"num_input_tokens_seen": 8985600,
|
| 1611 |
+
"step": 900
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 0.46613443214009787,
|
| 1615 |
+
"grad_norm": 2.090931980208863,
|
| 1616 |
+
"learning_rate": 8.775851227988656e-05,
|
| 1617 |
+
"loss": 0.0318,
|
| 1618 |
+
"num_input_tokens_seen": 9035520,
|
| 1619 |
+
"step": 905
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.46870976049446306,
|
| 1623 |
+
"grad_norm": 1.7373295104389102,
|
| 1624 |
+
"learning_rate": 8.759866980070963e-05,
|
| 1625 |
+
"loss": 0.0635,
|
| 1626 |
+
"num_input_tokens_seen": 9085440,
|
| 1627 |
+
"step": 910
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.47128508884882825,
|
| 1631 |
+
"grad_norm": 0.1557668148789241,
|
| 1632 |
+
"learning_rate": 8.743793810744654e-05,
|
| 1633 |
+
"loss": 0.0035,
|
| 1634 |
+
"num_input_tokens_seen": 9135360,
|
| 1635 |
+
"step": 915
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 0.4738604172031934,
|
| 1639 |
+
"grad_norm": 0.5241949871459053,
|
| 1640 |
+
"learning_rate": 8.727632100142551e-05,
|
| 1641 |
+
"loss": 0.0047,
|
| 1642 |
+
"num_input_tokens_seen": 9185280,
|
| 1643 |
+
"step": 920
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 0.4764357455575586,
|
| 1647 |
+
"grad_norm": 2.329131853129593,
|
| 1648 |
+
"learning_rate": 8.711382230491493e-05,
|
| 1649 |
+
"loss": 0.0194,
|
| 1650 |
+
"num_input_tokens_seen": 9235200,
|
| 1651 |
+
"step": 925
|
| 1652 |
+
},
|
| 1653 |
+
{
|
| 1654 |
+
"epoch": 0.47901107391192377,
|
| 1655 |
+
"grad_norm": 1.0158880737206768,
|
| 1656 |
+
"learning_rate": 8.695044586103296e-05,
|
| 1657 |
+
"loss": 0.0234,
|
| 1658 |
+
"num_input_tokens_seen": 9285120,
|
| 1659 |
+
"step": 930
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 0.48158640226628896,
|
| 1663 |
+
"grad_norm": 0.016729230547496875,
|
| 1664 |
+
"learning_rate": 8.678619553365659e-05,
|
| 1665 |
+
"loss": 0.0253,
|
| 1666 |
+
"num_input_tokens_seen": 9335040,
|
| 1667 |
+
"step": 935
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 0.48416173062065415,
|
| 1671 |
+
"grad_norm": 2.937987778998638,
|
| 1672 |
+
"learning_rate": 8.662107520733027e-05,
|
| 1673 |
+
"loss": 0.0191,
|
| 1674 |
+
"num_input_tokens_seen": 9384960,
|
| 1675 |
+
"step": 940
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.4867370589750193,
|
| 1679 |
+
"grad_norm": 0.20640046184009653,
|
| 1680 |
+
"learning_rate": 8.64550887871741e-05,
|
| 1681 |
+
"loss": 0.0469,
|
| 1682 |
+
"num_input_tokens_seen": 9434880,
|
| 1683 |
+
"step": 945
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.4893123873293845,
|
| 1687 |
+
"grad_norm": 1.9952606902760353,
|
| 1688 |
+
"learning_rate": 8.628824019879137e-05,
|
| 1689 |
+
"loss": 0.0415,
|
| 1690 |
+
"num_input_tokens_seen": 9484800,
|
| 1691 |
+
"step": 950
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 0.4893123873293845,
|
| 1695 |
+
"eval_loss": 0.007216573692858219,
|
| 1696 |
+
"eval_runtime": 19.2716,
|
| 1697 |
+
"eval_samples_per_second": 3.113,
|
| 1698 |
+
"eval_steps_per_second": 0.778,
|
| 1699 |
+
"num_input_tokens_seen": 9484800,
|
| 1700 |
+
"step": 950
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 0.49188771568374967,
|
| 1704 |
+
"grad_norm": 3.7704976412587468,
|
| 1705 |
+
"learning_rate": 8.612053338817581e-05,
|
| 1706 |
+
"loss": 0.0393,
|
| 1707 |
+
"num_input_tokens_seen": 9534720,
|
| 1708 |
+
"step": 955
|
| 1709 |
+
},
|
| 1710 |
+
{
|
| 1711 |
+
"epoch": 0.49446304403811486,
|
| 1712 |
+
"grad_norm": 0.2205817340051924,
|
| 1713 |
+
"learning_rate": 8.595197232161824e-05,
|
| 1714 |
+
"loss": 0.0102,
|
| 1715 |
+
"num_input_tokens_seen": 9584640,
|
| 1716 |
+
"step": 960
|
| 1717 |
+
},
|
| 1718 |
+
{
|
| 1719 |
+
"epoch": 0.49703837239248005,
|
| 1720 |
+
"grad_norm": 1.9646554268692766,
|
| 1721 |
+
"learning_rate": 8.578256098561275e-05,
|
| 1722 |
+
"loss": 0.0087,
|
| 1723 |
+
"num_input_tokens_seen": 9634560,
|
| 1724 |
+
"step": 965
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.49961370074684525,
|
| 1728 |
+
"grad_norm": 2.08637430640156,
|
| 1729 |
+
"learning_rate": 8.561230338676239e-05,
|
| 1730 |
+
"loss": 0.0097,
|
| 1731 |
+
"num_input_tokens_seen": 9684480,
|
| 1732 |
+
"step": 970
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.5021890291012104,
|
| 1736 |
+
"grad_norm": 0.018874732349478156,
|
| 1737 |
+
"learning_rate": 8.544120355168451e-05,
|
| 1738 |
+
"loss": 0.0013,
|
| 1739 |
+
"num_input_tokens_seen": 9734400,
|
| 1740 |
+
"step": 975
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 0.5047643574555756,
|
| 1744 |
+
"grad_norm": 0.007747713318690538,
|
| 1745 |
+
"learning_rate": 8.526926552691544e-05,
|
| 1746 |
+
"loss": 0.0268,
|
| 1747 |
+
"num_input_tokens_seen": 9784320,
|
| 1748 |
+
"step": 980
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 0.5073396858099408,
|
| 1752 |
+
"grad_norm": 4.2765032076143585,
|
| 1753 |
+
"learning_rate": 8.509649337881483e-05,
|
| 1754 |
+
"loss": 0.0418,
|
| 1755 |
+
"num_input_tokens_seen": 9834240,
|
| 1756 |
+
"step": 985
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 0.509915014164306,
|
| 1760 |
+
"grad_norm": 0.016725809588179666,
|
| 1761 |
+
"learning_rate": 8.492289119346943e-05,
|
| 1762 |
+
"loss": 0.0011,
|
| 1763 |
+
"num_input_tokens_seen": 9884160,
|
| 1764 |
+
"step": 990
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 0.5124903425186711,
|
| 1768 |
+
"grad_norm": 1.5189362175317798,
|
| 1769 |
+
"learning_rate": 8.474846307659658e-05,
|
| 1770 |
+
"loss": 0.0058,
|
| 1771 |
+
"num_input_tokens_seen": 9934080,
|
| 1772 |
+
"step": 995
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 0.5150656708730363,
|
| 1776 |
+
"grad_norm": 0.5873213891971129,
|
| 1777 |
+
"learning_rate": 8.457321315344694e-05,
|
| 1778 |
+
"loss": 0.0547,
|
| 1779 |
+
"num_input_tokens_seen": 9984000,
|
| 1780 |
+
"step": 1000
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.5150656708730363,
|
| 1784 |
+
"eval_loss": 0.0007357922149822116,
|
| 1785 |
+
"eval_runtime": 19.1915,
|
| 1786 |
+
"eval_samples_per_second": 3.126,
|
| 1787 |
+
"eval_steps_per_second": 0.782,
|
| 1788 |
+
"num_input_tokens_seen": 9984000,
|
| 1789 |
+
"step": 1000
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 0.5176409992274015,
|
| 1793 |
+
"grad_norm": 0.06909777020752475,
|
| 1794 |
+
"learning_rate": 8.439714556870704e-05,
|
| 1795 |
+
"loss": 0.0042,
|
| 1796 |
+
"num_input_tokens_seen": 10033920,
|
| 1797 |
+
"step": 1005
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"epoch": 0.5202163275817667,
|
| 1801 |
+
"grad_norm": 0.15306455110846962,
|
| 1802 |
+
"learning_rate": 8.422026448640124e-05,
|
| 1803 |
+
"loss": 0.0419,
|
| 1804 |
+
"num_input_tokens_seen": 10083840,
|
| 1805 |
+
"step": 1010
|
| 1806 |
+
},
|
| 1807 |
+
{
|
| 1808 |
+
"epoch": 0.5227916559361319,
|
| 1809 |
+
"grad_norm": 0.06864588320363613,
|
| 1810 |
+
"learning_rate": 8.40425740897932e-05,
|
| 1811 |
+
"loss": 0.0313,
|
| 1812 |
+
"num_input_tokens_seen": 10133760,
|
| 1813 |
+
"step": 1015
|
| 1814 |
+
},
|
| 1815 |
+
{
|
| 1816 |
+
"epoch": 0.525366984290497,
|
| 1817 |
+
"grad_norm": 0.041539288845711536,
|
| 1818 |
+
"learning_rate": 8.386407858128706e-05,
|
| 1819 |
+
"loss": 0.005,
|
| 1820 |
+
"num_input_tokens_seen": 10183680,
|
| 1821 |
+
"step": 1020
|
| 1822 |
+
},
|
| 1823 |
+
{
|
| 1824 |
+
"epoch": 0.5279423126448622,
|
| 1825 |
+
"grad_norm": 3.866462743446183,
|
| 1826 |
+
"learning_rate": 8.368478218232787e-05,
|
| 1827 |
+
"loss": 0.0171,
|
| 1828 |
+
"num_input_tokens_seen": 10233600,
|
| 1829 |
+
"step": 1025
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.5305176409992274,
|
| 1833 |
+
"grad_norm": 0.27036285469370147,
|
| 1834 |
+
"learning_rate": 8.350468913330192e-05,
|
| 1835 |
+
"loss": 0.0019,
|
| 1836 |
+
"num_input_tokens_seen": 10283520,
|
| 1837 |
+
"step": 1030
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.5330929693535926,
|
| 1841 |
+
"grad_norm": 0.12025617159059553,
|
| 1842 |
+
"learning_rate": 8.33238036934364e-05,
|
| 1843 |
+
"loss": 0.0256,
|
| 1844 |
+
"num_input_tokens_seen": 10333440,
|
| 1845 |
+
"step": 1035
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 0.5356682977079578,
|
| 1849 |
+
"grad_norm": 0.013500323699360073,
|
| 1850 |
+
"learning_rate": 8.31421301406986e-05,
|
| 1851 |
+
"loss": 0.0019,
|
| 1852 |
+
"num_input_tokens_seen": 10383360,
|
| 1853 |
+
"step": 1040
|
| 1854 |
+
},
|
| 1855 |
+
{
|
| 1856 |
+
"epoch": 0.5382436260623229,
|
| 1857 |
+
"grad_norm": 0.10118623585384687,
|
| 1858 |
+
"learning_rate": 8.29596727716949e-05,
|
| 1859 |
+
"loss": 0.018,
|
| 1860 |
+
"num_input_tokens_seen": 10433280,
|
| 1861 |
+
"step": 1045
|
| 1862 |
+
},
|
| 1863 |
+
{
|
| 1864 |
+
"epoch": 0.5408189544166881,
|
| 1865 |
+
"grad_norm": 0.23565575572231393,
|
| 1866 |
+
"learning_rate": 8.277643590156894e-05,
|
| 1867 |
+
"loss": 0.0007,
|
| 1868 |
+
"num_input_tokens_seen": 10483200,
|
| 1869 |
+
"step": 1050
|
| 1870 |
+
},
|
| 1871 |
+
{
|
| 1872 |
+
"epoch": 0.5408189544166881,
|
| 1873 |
+
"eval_loss": 0.05677948147058487,
|
| 1874 |
+
"eval_runtime": 19.0806,
|
| 1875 |
+
"eval_samples_per_second": 3.145,
|
| 1876 |
+
"eval_steps_per_second": 0.786,
|
| 1877 |
+
"num_input_tokens_seen": 10483200,
|
| 1878 |
+
"step": 1050
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.5433942827710533,
|
| 1882 |
+
"grad_norm": 6.026032426572499,
|
| 1883 |
+
"learning_rate": 8.259242386389973e-05,
|
| 1884 |
+
"loss": 0.0436,
|
| 1885 |
+
"num_input_tokens_seen": 10533120,
|
| 1886 |
+
"step": 1055
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.5459696111254185,
|
| 1890 |
+
"grad_norm": 9.168641384066685,
|
| 1891 |
+
"learning_rate": 8.240764101059912e-05,
|
| 1892 |
+
"loss": 0.0124,
|
| 1893 |
+
"num_input_tokens_seen": 10583040,
|
| 1894 |
+
"step": 1060
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 0.5485449394797837,
|
| 1898 |
+
"grad_norm": 0.590720691746965,
|
| 1899 |
+
"learning_rate": 8.222209171180883e-05,
|
| 1900 |
+
"loss": 0.0155,
|
| 1901 |
+
"num_input_tokens_seen": 10632960,
|
| 1902 |
+
"step": 1065
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"epoch": 0.5511202678341488,
|
| 1906 |
+
"grad_norm": 8.790606959622112,
|
| 1907 |
+
"learning_rate": 8.203578035579715e-05,
|
| 1908 |
+
"loss": 0.0206,
|
| 1909 |
+
"num_input_tokens_seen": 10682880,
|
| 1910 |
+
"step": 1070
|
| 1911 |
+
},
|
| 1912 |
+
{
|
| 1913 |
+
"epoch": 0.553695596188514,
|
| 1914 |
+
"grad_norm": 0.11045279468612182,
|
| 1915 |
+
"learning_rate": 8.184871134885513e-05,
|
| 1916 |
+
"loss": 0.0237,
|
| 1917 |
+
"num_input_tokens_seen": 10732800,
|
| 1918 |
+
"step": 1075
|
| 1919 |
+
},
|
| 1920 |
+
{
|
| 1921 |
+
"epoch": 0.5562709245428792,
|
| 1922 |
+
"grad_norm": 0.004997644768678168,
|
| 1923 |
+
"learning_rate": 8.166088911519235e-05,
|
| 1924 |
+
"loss": 0.0018,
|
| 1925 |
+
"num_input_tokens_seen": 10782720,
|
| 1926 |
+
"step": 1080
|
| 1927 |
+
},
|
| 1928 |
+
{
|
| 1929 |
+
"epoch": 0.5588462528972444,
|
| 1930 |
+
"grad_norm": 2.249874392718823,
|
| 1931 |
+
"learning_rate": 8.147231809683236e-05,
|
| 1932 |
+
"loss": 0.0296,
|
| 1933 |
+
"num_input_tokens_seen": 10832640,
|
| 1934 |
+
"step": 1085
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.5614215812516096,
|
| 1938 |
+
"grad_norm": 0.12969877743086014,
|
| 1939 |
+
"learning_rate": 8.128300275350756e-05,
|
| 1940 |
+
"loss": 0.0006,
|
| 1941 |
+
"num_input_tokens_seen": 10882560,
|
| 1942 |
+
"step": 1090
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.5639969096059748,
|
| 1946 |
+
"grad_norm": 0.01592203313456207,
|
| 1947 |
+
"learning_rate": 8.109294756255375e-05,
|
| 1948 |
+
"loss": 0.0004,
|
| 1949 |
+
"num_input_tokens_seen": 10932480,
|
| 1950 |
+
"step": 1095
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 0.56657223796034,
|
| 1954 |
+
"grad_norm": 0.048956329384631035,
|
| 1955 |
+
"learning_rate": 8.090215701880419e-05,
|
| 1956 |
+
"loss": 0.0056,
|
| 1957 |
+
"num_input_tokens_seen": 10982400,
|
| 1958 |
+
"step": 1100
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 0.56657223796034,
|
| 1962 |
+
"eval_loss": 0.0003656313638202846,
|
| 1963 |
+
"eval_runtime": 19.3051,
|
| 1964 |
+
"eval_samples_per_second": 3.108,
|
| 1965 |
+
"eval_steps_per_second": 0.777,
|
| 1966 |
+
"num_input_tokens_seen": 10982400,
|
| 1967 |
+
"step": 1100
|
| 1968 |
+
},
|
| 1969 |
+
{
|
| 1970 |
+
"epoch": 0.5691475663147051,
|
| 1971 |
+
"grad_norm": 0.020540774848059624,
|
| 1972 |
+
"learning_rate": 8.07106356344834e-05,
|
| 1973 |
+
"loss": 0.0005,
|
| 1974 |
+
"num_input_tokens_seen": 11032320,
|
| 1975 |
+
"step": 1105
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 0.5717228946690703,
|
| 1979 |
+
"grad_norm": 0.049646985711380834,
|
| 1980 |
+
"learning_rate": 8.051838793910038e-05,
|
| 1981 |
+
"loss": 0.0008,
|
| 1982 |
+
"num_input_tokens_seen": 11082240,
|
| 1983 |
+
"step": 1110
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.5742982230234355,
|
| 1987 |
+
"grad_norm": 0.02410302194894019,
|
| 1988 |
+
"learning_rate": 8.032541847934146e-05,
|
| 1989 |
+
"loss": 0.001,
|
| 1990 |
+
"num_input_tokens_seen": 11132160,
|
| 1991 |
+
"step": 1115
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.5768735513778007,
|
| 1995 |
+
"grad_norm": 0.03404095257940449,
|
| 1996 |
+
"learning_rate": 8.013173181896283e-05,
|
| 1997 |
+
"loss": 0.0219,
|
| 1998 |
+
"num_input_tokens_seen": 11182080,
|
| 1999 |
+
"step": 1120
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 0.5794488797321659,
|
| 2003 |
+
"grad_norm": 0.0005611895912019751,
|
| 2004 |
+
"learning_rate": 7.993733253868256e-05,
|
| 2005 |
+
"loss": 0.0,
|
| 2006 |
+
"num_input_tokens_seen": 11232000,
|
| 2007 |
+
"step": 1125
|
| 2008 |
+
},
|
| 2009 |
+
{
|
| 2010 |
+
"epoch": 0.582024208086531,
|
| 2011 |
+
"grad_norm": 0.03678453984585236,
|
| 2012 |
+
"learning_rate": 7.974222523607236e-05,
|
| 2013 |
+
"loss": 0.0,
|
| 2014 |
+
"num_input_tokens_seen": 11281920,
|
| 2015 |
+
"step": 1130
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 0.5845995364408962,
|
| 2019 |
+
"grad_norm": 0.002820815423322981,
|
| 2020 |
+
"learning_rate": 7.954641452544865e-05,
|
| 2021 |
+
"loss": 0.0001,
|
| 2022 |
+
"num_input_tokens_seen": 11331840,
|
| 2023 |
+
"step": 1135
|
| 2024 |
+
},
|
| 2025 |
+
{
|
| 2026 |
+
"epoch": 0.5871748647952614,
|
| 2027 |
+
"grad_norm": 0.003003123169716893,
|
| 2028 |
+
"learning_rate": 7.934990503776363e-05,
|
| 2029 |
+
"loss": 0.0,
|
| 2030 |
+
"num_input_tokens_seen": 11381760,
|
| 2031 |
+
"step": 1140
|
| 2032 |
+
},
|
| 2033 |
+
{
|
| 2034 |
+
"epoch": 0.5897501931496266,
|
| 2035 |
+
"grad_norm": 0.020018776153065302,
|
| 2036 |
+
"learning_rate": 7.915270142049566e-05,
|
| 2037 |
+
"loss": 0.0001,
|
| 2038 |
+
"num_input_tokens_seen": 11431680,
|
| 2039 |
+
"step": 1145
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.5923255215039918,
|
| 2043 |
+
"grad_norm": 0.0005407295497242103,
|
| 2044 |
+
"learning_rate": 7.89548083375394e-05,
|
| 2045 |
+
"loss": 0.0127,
|
| 2046 |
+
"num_input_tokens_seen": 11481600,
|
| 2047 |
+
"step": 1150
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.5923255215039918,
|
| 2051 |
+
"eval_loss": 1.5566551155643538e-05,
|
| 2052 |
+
"eval_runtime": 19.3425,
|
| 2053 |
+
"eval_samples_per_second": 3.102,
|
| 2054 |
+
"eval_steps_per_second": 0.775,
|
| 2055 |
+
"num_input_tokens_seen": 11481600,
|
| 2056 |
+
"step": 1150
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"epoch": 0.5949008498583569,
|
| 2060 |
+
"grad_norm": 0.00046149068264744525,
|
| 2061 |
+
"learning_rate": 7.875623046909544e-05,
|
| 2062 |
+
"loss": 0.0003,
|
| 2063 |
+
"num_input_tokens_seen": 11531520,
|
| 2064 |
+
"step": 1155
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 0.5974761782127221,
|
| 2068 |
+
"grad_norm": 0.0007891686440962277,
|
| 2069 |
+
"learning_rate": 7.855697251155967e-05,
|
| 2070 |
+
"loss": 0.0,
|
| 2071 |
+
"num_input_tokens_seen": 11581440,
|
| 2072 |
+
"step": 1160
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 0.6000515065670873,
|
| 2076 |
+
"grad_norm": 6.055667685356653,
|
| 2077 |
+
"learning_rate": 7.835703917741212e-05,
|
| 2078 |
+
"loss": 0.0414,
|
| 2079 |
+
"num_input_tokens_seen": 11631360,
|
| 2080 |
+
"step": 1165
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 0.6026268349214525,
|
| 2084 |
+
"grad_norm": 3.9367530795344696,
|
| 2085 |
+
"learning_rate": 7.81564351951057e-05,
|
| 2086 |
+
"loss": 0.02,
|
| 2087 |
+
"num_input_tokens_seen": 11681280,
|
| 2088 |
+
"step": 1170
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.6052021632758177,
|
| 2092 |
+
"grad_norm": 0.014148361508830113,
|
| 2093 |
+
"learning_rate": 7.795516530895414e-05,
|
| 2094 |
+
"loss": 0.0002,
|
| 2095 |
+
"num_input_tokens_seen": 11731200,
|
| 2096 |
+
"step": 1175
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.6077774916301828,
|
| 2100 |
+
"grad_norm": 0.12219412030414617,
|
| 2101 |
+
"learning_rate": 7.775323427901993e-05,
|
| 2102 |
+
"loss": 0.0026,
|
| 2103 |
+
"num_input_tokens_seen": 11781120,
|
| 2104 |
+
"step": 1180
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 0.610352819984548,
|
| 2108 |
+
"grad_norm": 5.108798582105785,
|
| 2109 |
+
"learning_rate": 7.755064688100171e-05,
|
| 2110 |
+
"loss": 0.0168,
|
| 2111 |
+
"num_input_tokens_seen": 11831040,
|
| 2112 |
+
"step": 1185
|
| 2113 |
+
},
|
| 2114 |
+
{
|
| 2115 |
+
"epoch": 0.6129281483389132,
|
| 2116 |
+
"grad_norm": 0.004932933139070266,
|
| 2117 |
+
"learning_rate": 7.734740790612136e-05,
|
| 2118 |
+
"loss": 0.0037,
|
| 2119 |
+
"num_input_tokens_seen": 11880960,
|
| 2120 |
+
"step": 1190
|
| 2121 |
+
},
|
| 2122 |
+
{
|
| 2123 |
+
"epoch": 0.6155034766932784,
|
| 2124 |
+
"grad_norm": 0.022109282268939556,
|
| 2125 |
+
"learning_rate": 7.714352216101055e-05,
|
| 2126 |
+
"loss": 0.0007,
|
| 2127 |
+
"num_input_tokens_seen": 11930880,
|
| 2128 |
+
"step": 1195
|
| 2129 |
+
},
|
| 2130 |
+
{
|
| 2131 |
+
"epoch": 0.6180788050476436,
|
| 2132 |
+
"grad_norm": 0.037426371340479145,
|
| 2133 |
+
"learning_rate": 7.693899446759727e-05,
|
| 2134 |
+
"loss": 0.0038,
|
| 2135 |
+
"num_input_tokens_seen": 11980800,
|
| 2136 |
+
"step": 1200
|
| 2137 |
+
},
|
| 2138 |
+
{
|
| 2139 |
+
"epoch": 0.6180788050476436,
|
| 2140 |
+
"eval_loss": 0.0022160401567816734,
|
| 2141 |
+
"eval_runtime": 18.725,
|
| 2142 |
+
"eval_samples_per_second": 3.204,
|
| 2143 |
+
"eval_steps_per_second": 0.801,
|
| 2144 |
+
"num_input_tokens_seen": 11980800,
|
| 2145 |
+
"step": 1200
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.6185938707185166,
|
| 2149 |
+
"num_input_tokens_seen": 11990784,
|
| 2150 |
+
"step": 1201,
|
| 2151 |
+
"total_flos": 791107965943808.0,
|
| 2152 |
+
"train_loss": 6.11944748344618e-08,
|
| 2153 |
+
"train_runtime": 102.3684,
|
| 2154 |
+
"train_samples_per_second": 281.337,
|
| 2155 |
+
"train_steps_per_second": 11.722
|
| 2156 |
+
}
|
| 2157 |
+
],
|
| 2158 |
+
"logging_steps": 5,
|
| 2159 |
+
"max_steps": 1200,
|
| 2160 |
+
"num_input_tokens_seen": 11990784,
|
| 2161 |
+
"num_train_epochs": 1,
|
| 2162 |
+
"save_steps": 50,
|
| 2163 |
+
"stateful_callbacks": {
|
| 2164 |
+
"TrainerControl": {
|
| 2165 |
+
"args": {
|
| 2166 |
+
"should_epoch_stop": false,
|
| 2167 |
+
"should_evaluate": false,
|
| 2168 |
+
"should_log": false,
|
| 2169 |
+
"should_save": true,
|
| 2170 |
+
"should_training_stop": true
|
| 2171 |
+
},
|
| 2172 |
+
"attributes": {}
|
| 2173 |
+
}
|
| 2174 |
+
},
|
| 2175 |
+
"total_flos": 791107965943808.0,
|
| 2176 |
+
"train_batch_size": 1,
|
| 2177 |
+
"trial_name": null,
|
| 2178 |
+
"trial_params": null
|
| 2179 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|