End of training
Browse files- README.md +16 -3
- all_results.json +16 -0
- eval_results.json +10 -0
- train_results.json +9 -0
- trainer_state.json +2342 -0
README.md
CHANGED
|
@@ -2,11 +2,24 @@
|
|
| 2 |
library_name: transformers
|
| 3 |
tags:
|
| 4 |
- generated_from_trainer
|
|
|
|
|
|
|
| 5 |
metrics:
|
| 6 |
- accuracy
|
| 7 |
model-index:
|
| 8 |
- name: T5LA
|
| 9 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
---
|
| 11 |
|
| 12 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -15,10 +28,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/uoy/llm_training/runs/elf928gg)
|
| 16 |
# T5LA
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [](https://huggingface.co/) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Accuracy: 0.0322
|
| 21 |
- Loss: 5.5470
|
|
|
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 2 |
library_name: transformers
|
| 3 |
tags:
|
| 4 |
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- HuggingFaceFW/fineweb
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
| 9 |
model-index:
|
| 10 |
- name: T5LA
|
| 11 |
+
results:
|
| 12 |
+
- task:
|
| 13 |
+
name: Causal Language Modeling
|
| 14 |
+
type: text-generation
|
| 15 |
+
dataset:
|
| 16 |
+
name: HuggingFaceFW/fineweb sample-10BT
|
| 17 |
+
type: HuggingFaceFW/fineweb
|
| 18 |
+
args: sample-10BT
|
| 19 |
+
metrics:
|
| 20 |
+
- name: Accuracy
|
| 21 |
+
type: accuracy
|
| 22 |
+
value: 0.03222989830774154
|
| 23 |
---
|
| 24 |
|
| 25 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 28 |
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/uoy/llm_training/runs/elf928gg)
|
| 29 |
# T5LA
|
| 30 |
|
| 31 |
+
This model is a fine-tuned version of [](https://huggingface.co/) on the HuggingFaceFW/fineweb sample-10BT dataset.
|
| 32 |
It achieves the following results on the evaluation set:
|
|
|
|
| 33 |
- Loss: 5.5470
|
| 34 |
+
- Accuracy: 0.0322
|
| 35 |
|
| 36 |
## Model description
|
| 37 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.00001,
|
| 3 |
+
"eval_accuracy": 0.03222989830774154,
|
| 4 |
+
"eval_loss": 5.5469770431518555,
|
| 5 |
+
"eval_runtime": 110.5546,
|
| 6 |
+
"eval_samples": 10000,
|
| 7 |
+
"eval_samples_per_second": 32.491,
|
| 8 |
+
"eval_steps_per_second": 2.035,
|
| 9 |
+
"perplexity": 256.46111204397334,
|
| 10 |
+
"total_flos": 9.182126159167488e+17,
|
| 11 |
+
"train_loss": 5.625401986489168e-05,
|
| 12 |
+
"train_runtime": 26.8473,
|
| 13 |
+
"train_samples": 1000000,
|
| 14 |
+
"train_samples_per_second": 59596.412,
|
| 15 |
+
"train_steps_per_second": 3724.776
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.00001,
|
| 3 |
+
"eval_accuracy": 0.03222989830774154,
|
| 4 |
+
"eval_loss": 5.5469770431518555,
|
| 5 |
+
"eval_runtime": 110.5546,
|
| 6 |
+
"eval_samples": 10000,
|
| 7 |
+
"eval_samples_per_second": 32.491,
|
| 8 |
+
"eval_steps_per_second": 2.035,
|
| 9 |
+
"perplexity": 256.46111204397334
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.00001,
|
| 3 |
+
"total_flos": 9.182126159167488e+17,
|
| 4 |
+
"train_loss": 5.625401986489168e-05,
|
| 5 |
+
"train_runtime": 26.8473,
|
| 6 |
+
"train_samples": 1000000,
|
| 7 |
+
"train_samples_per_second": 59596.412,
|
| 8 |
+
"train_steps_per_second": 3724.776
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2342 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 5.546974182128906,
|
| 3 |
+
"best_model_checkpoint": "/users/hr1171/scratch/T5LA/checkpoint-100000",
|
| 4 |
+
"epoch": 1.00001,
|
| 5 |
+
"eval_steps": 1000,
|
| 6 |
+
"global_step": 100001,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.005,
|
| 13 |
+
"grad_norm": 0.06483318656682968,
|
| 14 |
+
"learning_rate": 4.975e-05,
|
| 15 |
+
"loss": 10.0248,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.01,
|
| 20 |
+
"grad_norm": 0.07472355663776398,
|
| 21 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 22 |
+
"loss": 9.4056,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.01,
|
| 27 |
+
"eval_accuracy": 0.043492843878108624,
|
| 28 |
+
"eval_loss": 9.121541023254395,
|
| 29 |
+
"eval_runtime": 110.4799,
|
| 30 |
+
"eval_samples_per_second": 32.513,
|
| 31 |
+
"eval_steps_per_second": 2.037,
|
| 32 |
+
"step": 1000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.015,
|
| 36 |
+
"grad_norm": 0.08399348706007004,
|
| 37 |
+
"learning_rate": 4.9250000000000004e-05,
|
| 38 |
+
"loss": 8.8698,
|
| 39 |
+
"step": 1500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.02,
|
| 43 |
+
"grad_norm": 0.08570190519094467,
|
| 44 |
+
"learning_rate": 4.9e-05,
|
| 45 |
+
"loss": 8.4062,
|
| 46 |
+
"step": 2000
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.02,
|
| 50 |
+
"eval_accuracy": 0.04429728167514647,
|
| 51 |
+
"eval_loss": 8.193856239318848,
|
| 52 |
+
"eval_runtime": 112.0891,
|
| 53 |
+
"eval_samples_per_second": 32.046,
|
| 54 |
+
"eval_steps_per_second": 2.007,
|
| 55 |
+
"step": 2000
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.025,
|
| 59 |
+
"grad_norm": 0.08781777322292328,
|
| 60 |
+
"learning_rate": 4.875e-05,
|
| 61 |
+
"loss": 8.0175,
|
| 62 |
+
"step": 2500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.03,
|
| 66 |
+
"grad_norm": 0.07474139332771301,
|
| 67 |
+
"learning_rate": 4.85e-05,
|
| 68 |
+
"loss": 7.7307,
|
| 69 |
+
"step": 3000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.03,
|
| 73 |
+
"eval_accuracy": 0.04438749518317016,
|
| 74 |
+
"eval_loss": 7.602397918701172,
|
| 75 |
+
"eval_runtime": 112.6714,
|
| 76 |
+
"eval_samples_per_second": 31.88,
|
| 77 |
+
"eval_steps_per_second": 1.997,
|
| 78 |
+
"step": 3000
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.035,
|
| 82 |
+
"grad_norm": 0.07133983075618744,
|
| 83 |
+
"learning_rate": 4.825e-05,
|
| 84 |
+
"loss": 7.5203,
|
| 85 |
+
"step": 3500
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.04,
|
| 89 |
+
"grad_norm": 0.04410657659173012,
|
| 90 |
+
"learning_rate": 4.8e-05,
|
| 91 |
+
"loss": 7.39,
|
| 92 |
+
"step": 4000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.04,
|
| 96 |
+
"eval_accuracy": 0.0443981085370553,
|
| 97 |
+
"eval_loss": 7.333784103393555,
|
| 98 |
+
"eval_runtime": 111.8336,
|
| 99 |
+
"eval_samples_per_second": 32.119,
|
| 100 |
+
"eval_steps_per_second": 2.012,
|
| 101 |
+
"step": 4000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.045,
|
| 105 |
+
"grad_norm": 0.03503479063510895,
|
| 106 |
+
"learning_rate": 4.775e-05,
|
| 107 |
+
"loss": 7.3043,
|
| 108 |
+
"step": 4500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.05,
|
| 112 |
+
"grad_norm": 0.039906181395053864,
|
| 113 |
+
"learning_rate": 4.75e-05,
|
| 114 |
+
"loss": 7.2546,
|
| 115 |
+
"step": 5000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.05,
|
| 119 |
+
"eval_accuracy": 0.04405276632986957,
|
| 120 |
+
"eval_loss": 7.245168685913086,
|
| 121 |
+
"eval_runtime": 113.4504,
|
| 122 |
+
"eval_samples_per_second": 31.661,
|
| 123 |
+
"eval_steps_per_second": 1.983,
|
| 124 |
+
"step": 5000
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 0.055,
|
| 128 |
+
"grad_norm": 0.04767516627907753,
|
| 129 |
+
"learning_rate": 4.7249999999999997e-05,
|
| 130 |
+
"loss": 7.2409,
|
| 131 |
+
"step": 5500
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.06,
|
| 135 |
+
"grad_norm": 0.07150296866893768,
|
| 136 |
+
"learning_rate": 4.7e-05,
|
| 137 |
+
"loss": 7.1985,
|
| 138 |
+
"step": 6000
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.06,
|
| 142 |
+
"eval_accuracy": 0.03687500952480477,
|
| 143 |
+
"eval_loss": 7.168196201324463,
|
| 144 |
+
"eval_runtime": 110.8688,
|
| 145 |
+
"eval_samples_per_second": 32.399,
|
| 146 |
+
"eval_steps_per_second": 2.029,
|
| 147 |
+
"step": 6000
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.065,
|
| 151 |
+
"grad_norm": 0.10620597004890442,
|
| 152 |
+
"learning_rate": 4.6750000000000005e-05,
|
| 153 |
+
"loss": 7.1475,
|
| 154 |
+
"step": 6500
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.07,
|
| 158 |
+
"grad_norm": 0.09856470674276352,
|
| 159 |
+
"learning_rate": 4.6500000000000005e-05,
|
| 160 |
+
"loss": 7.1009,
|
| 161 |
+
"step": 7000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.07,
|
| 165 |
+
"eval_accuracy": 0.03457599379091584,
|
| 166 |
+
"eval_loss": 7.071776866912842,
|
| 167 |
+
"eval_runtime": 115.3179,
|
| 168 |
+
"eval_samples_per_second": 31.149,
|
| 169 |
+
"eval_steps_per_second": 1.951,
|
| 170 |
+
"step": 7000
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.075,
|
| 174 |
+
"grad_norm": 0.12105035036802292,
|
| 175 |
+
"learning_rate": 4.6250000000000006e-05,
|
| 176 |
+
"loss": 7.0493,
|
| 177 |
+
"step": 7500
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.08,
|
| 181 |
+
"grad_norm": 0.1322629749774933,
|
| 182 |
+
"learning_rate": 4.600000000000001e-05,
|
| 183 |
+
"loss": 7.004,
|
| 184 |
+
"step": 8000
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.08,
|
| 188 |
+
"eval_accuracy": 0.03315856677269135,
|
| 189 |
+
"eval_loss": 6.977820873260498,
|
| 190 |
+
"eval_runtime": 110.4899,
|
| 191 |
+
"eval_samples_per_second": 32.51,
|
| 192 |
+
"eval_steps_per_second": 2.036,
|
| 193 |
+
"step": 8000
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.085,
|
| 197 |
+
"grad_norm": 0.16380475461483002,
|
| 198 |
+
"learning_rate": 4.575e-05,
|
| 199 |
+
"loss": 6.9535,
|
| 200 |
+
"step": 8500
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.09,
|
| 204 |
+
"grad_norm": 0.1642669439315796,
|
| 205 |
+
"learning_rate": 4.55e-05,
|
| 206 |
+
"loss": 6.9159,
|
| 207 |
+
"step": 9000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.09,
|
| 211 |
+
"eval_accuracy": 0.032452642670689945,
|
| 212 |
+
"eval_loss": 6.896429061889648,
|
| 213 |
+
"eval_runtime": 112.221,
|
| 214 |
+
"eval_samples_per_second": 32.008,
|
| 215 |
+
"eval_steps_per_second": 2.005,
|
| 216 |
+
"step": 9000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.095,
|
| 220 |
+
"grad_norm": 0.2253541350364685,
|
| 221 |
+
"learning_rate": 4.525e-05,
|
| 222 |
+
"loss": 6.8866,
|
| 223 |
+
"step": 9500
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 0.1,
|
| 227 |
+
"grad_norm": 0.20757266879081726,
|
| 228 |
+
"learning_rate": 4.5e-05,
|
| 229 |
+
"loss": 6.8548,
|
| 230 |
+
"step": 10000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.1,
|
| 234 |
+
"eval_accuracy": 0.03252571153012995,
|
| 235 |
+
"eval_loss": 6.830728054046631,
|
| 236 |
+
"eval_runtime": 112.9087,
|
| 237 |
+
"eval_samples_per_second": 31.813,
|
| 238 |
+
"eval_steps_per_second": 1.993,
|
| 239 |
+
"step": 10000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.105,
|
| 243 |
+
"grad_norm": 0.21346400678157806,
|
| 244 |
+
"learning_rate": 4.4750000000000004e-05,
|
| 245 |
+
"loss": 6.8302,
|
| 246 |
+
"step": 10500
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.11,
|
| 250 |
+
"grad_norm": 0.24308614432811737,
|
| 251 |
+
"learning_rate": 4.4500000000000004e-05,
|
| 252 |
+
"loss": 6.7833,
|
| 253 |
+
"step": 11000
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.11,
|
| 257 |
+
"eval_accuracy": 0.032619326754142475,
|
| 258 |
+
"eval_loss": 6.77016544342041,
|
| 259 |
+
"eval_runtime": 112.3436,
|
| 260 |
+
"eval_samples_per_second": 31.973,
|
| 261 |
+
"eval_steps_per_second": 2.003,
|
| 262 |
+
"step": 11000
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.115,
|
| 266 |
+
"grad_norm": 0.27737605571746826,
|
| 267 |
+
"learning_rate": 4.4250000000000005e-05,
|
| 268 |
+
"loss": 6.7775,
|
| 269 |
+
"step": 11500
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.12,
|
| 273 |
+
"grad_norm": 0.3234957158565521,
|
| 274 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 275 |
+
"loss": 6.7376,
|
| 276 |
+
"step": 12000
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.12,
|
| 280 |
+
"eval_accuracy": 0.03369168370246034,
|
| 281 |
+
"eval_loss": 6.716267108917236,
|
| 282 |
+
"eval_runtime": 111.859,
|
| 283 |
+
"eval_samples_per_second": 32.112,
|
| 284 |
+
"eval_steps_per_second": 2.011,
|
| 285 |
+
"step": 12000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.125,
|
| 289 |
+
"grad_norm": 0.27591678500175476,
|
| 290 |
+
"learning_rate": 4.375e-05,
|
| 291 |
+
"loss": 6.7188,
|
| 292 |
+
"step": 12500
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.13,
|
| 296 |
+
"grad_norm": 0.31338784098625183,
|
| 297 |
+
"learning_rate": 4.35005e-05,
|
| 298 |
+
"loss": 6.6821,
|
| 299 |
+
"step": 13000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.13,
|
| 303 |
+
"eval_accuracy": 0.03455980162280902,
|
| 304 |
+
"eval_loss": 6.661470413208008,
|
| 305 |
+
"eval_runtime": 120.7993,
|
| 306 |
+
"eval_samples_per_second": 29.735,
|
| 307 |
+
"eval_steps_per_second": 1.863,
|
| 308 |
+
"step": 13000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.135,
|
| 312 |
+
"grad_norm": 0.513124406337738,
|
| 313 |
+
"learning_rate": 4.32505e-05,
|
| 314 |
+
"loss": 6.6565,
|
| 315 |
+
"step": 13500
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 0.14,
|
| 319 |
+
"grad_norm": 0.40841469168663025,
|
| 320 |
+
"learning_rate": 4.30005e-05,
|
| 321 |
+
"loss": 6.6373,
|
| 322 |
+
"step": 14000
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 0.14,
|
| 326 |
+
"eval_accuracy": 0.03493249362654492,
|
| 327 |
+
"eval_loss": 6.608607292175293,
|
| 328 |
+
"eval_runtime": 111.8763,
|
| 329 |
+
"eval_samples_per_second": 32.107,
|
| 330 |
+
"eval_steps_per_second": 2.011,
|
| 331 |
+
"step": 14000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.145,
|
| 335 |
+
"grad_norm": 0.3946859538555145,
|
| 336 |
+
"learning_rate": 4.2750500000000003e-05,
|
| 337 |
+
"loss": 6.6188,
|
| 338 |
+
"step": 14500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.15,
|
| 342 |
+
"grad_norm": 0.48516473174095154,
|
| 343 |
+
"learning_rate": 4.2501000000000005e-05,
|
| 344 |
+
"loss": 6.5895,
|
| 345 |
+
"step": 15000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.15,
|
| 349 |
+
"eval_accuracy": 0.0343823681168318,
|
| 350 |
+
"eval_loss": 6.556947231292725,
|
| 351 |
+
"eval_runtime": 112.7205,
|
| 352 |
+
"eval_samples_per_second": 31.866,
|
| 353 |
+
"eval_steps_per_second": 1.996,
|
| 354 |
+
"step": 15000
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.155,
|
| 358 |
+
"grad_norm": 0.543297529220581,
|
| 359 |
+
"learning_rate": 4.22515e-05,
|
| 360 |
+
"loss": 6.5555,
|
| 361 |
+
"step": 15500
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.16,
|
| 365 |
+
"grad_norm": 0.46242478489875793,
|
| 366 |
+
"learning_rate": 4.200150000000001e-05,
|
| 367 |
+
"loss": 6.5421,
|
| 368 |
+
"step": 16000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.16,
|
| 372 |
+
"eval_accuracy": 0.035396487687420944,
|
| 373 |
+
"eval_loss": 6.511915683746338,
|
| 374 |
+
"eval_runtime": 111.3995,
|
| 375 |
+
"eval_samples_per_second": 32.244,
|
| 376 |
+
"eval_steps_per_second": 2.02,
|
| 377 |
+
"step": 16000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.165,
|
| 381 |
+
"grad_norm": 0.5675227642059326,
|
| 382 |
+
"learning_rate": 4.17515e-05,
|
| 383 |
+
"loss": 6.5233,
|
| 384 |
+
"step": 16500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.17,
|
| 388 |
+
"grad_norm": 0.5619395971298218,
|
| 389 |
+
"learning_rate": 4.15015e-05,
|
| 390 |
+
"loss": 6.5051,
|
| 391 |
+
"step": 17000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.17,
|
| 395 |
+
"eval_accuracy": 0.035478945282990115,
|
| 396 |
+
"eval_loss": 6.467820644378662,
|
| 397 |
+
"eval_runtime": 111.7569,
|
| 398 |
+
"eval_samples_per_second": 32.141,
|
| 399 |
+
"eval_steps_per_second": 2.013,
|
| 400 |
+
"step": 17000
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.175,
|
| 404 |
+
"grad_norm": 0.4741845726966858,
|
| 405 |
+
"learning_rate": 4.1252000000000004e-05,
|
| 406 |
+
"loss": 6.491,
|
| 407 |
+
"step": 17500
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.18,
|
| 411 |
+
"grad_norm": 0.5824158191680908,
|
| 412 |
+
"learning_rate": 4.1002000000000005e-05,
|
| 413 |
+
"loss": 6.4391,
|
| 414 |
+
"step": 18000
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.18,
|
| 418 |
+
"eval_accuracy": 0.03599301260322167,
|
| 419 |
+
"eval_loss": 6.432408332824707,
|
| 420 |
+
"eval_runtime": 113.2507,
|
| 421 |
+
"eval_samples_per_second": 31.717,
|
| 422 |
+
"eval_steps_per_second": 1.987,
|
| 423 |
+
"step": 18000
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.185,
|
| 427 |
+
"grad_norm": 0.740391194820404,
|
| 428 |
+
"learning_rate": 4.0752e-05,
|
| 429 |
+
"loss": 6.4602,
|
| 430 |
+
"step": 18500
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.19,
|
| 434 |
+
"grad_norm": 0.55320805311203,
|
| 435 |
+
"learning_rate": 4.0502e-05,
|
| 436 |
+
"loss": 6.4242,
|
| 437 |
+
"step": 19000
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.19,
|
| 441 |
+
"eval_accuracy": 0.03552126262989112,
|
| 442 |
+
"eval_loss": 6.40146541595459,
|
| 443 |
+
"eval_runtime": 111.8361,
|
| 444 |
+
"eval_samples_per_second": 32.118,
|
| 445 |
+
"eval_steps_per_second": 2.012,
|
| 446 |
+
"step": 19000
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.195,
|
| 450 |
+
"grad_norm": 0.7065308094024658,
|
| 451 |
+
"learning_rate": 4.0252e-05,
|
| 452 |
+
"loss": 6.4047,
|
| 453 |
+
"step": 19500
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 0.2,
|
| 457 |
+
"grad_norm": 0.6466010212898254,
|
| 458 |
+
"learning_rate": 4.0003e-05,
|
| 459 |
+
"loss": 6.3889,
|
| 460 |
+
"step": 20000
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.2,
|
| 464 |
+
"eval_accuracy": 0.03732308355485308,
|
| 465 |
+
"eval_loss": 6.355311393737793,
|
| 466 |
+
"eval_runtime": 112.8526,
|
| 467 |
+
"eval_samples_per_second": 31.829,
|
| 468 |
+
"eval_steps_per_second": 1.994,
|
| 469 |
+
"step": 20000
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.205,
|
| 473 |
+
"grad_norm": 0.5905252695083618,
|
| 474 |
+
"learning_rate": 3.9753000000000004e-05,
|
| 475 |
+
"loss": 6.3693,
|
| 476 |
+
"step": 20500
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.21,
|
| 480 |
+
"grad_norm": 0.5333964824676514,
|
| 481 |
+
"learning_rate": 3.9503000000000004e-05,
|
| 482 |
+
"loss": 6.3631,
|
| 483 |
+
"step": 21000
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 0.21,
|
| 487 |
+
"eval_accuracy": 0.036739757297089004,
|
| 488 |
+
"eval_loss": 6.328461647033691,
|
| 489 |
+
"eval_runtime": 112.8636,
|
| 490 |
+
"eval_samples_per_second": 31.826,
|
| 491 |
+
"eval_steps_per_second": 1.994,
|
| 492 |
+
"step": 21000
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.215,
|
| 496 |
+
"grad_norm": 0.5893262028694153,
|
| 497 |
+
"learning_rate": 3.9253e-05,
|
| 498 |
+
"loss": 6.3469,
|
| 499 |
+
"step": 21500
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.22,
|
| 503 |
+
"grad_norm": 0.596720278263092,
|
| 504 |
+
"learning_rate": 3.9003e-05,
|
| 505 |
+
"loss": 6.3296,
|
| 506 |
+
"step": 22000
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.22,
|
| 510 |
+
"eval_accuracy": 0.03686983891650175,
|
| 511 |
+
"eval_loss": 6.3014631271362305,
|
| 512 |
+
"eval_runtime": 111.598,
|
| 513 |
+
"eval_samples_per_second": 32.187,
|
| 514 |
+
"eval_steps_per_second": 2.016,
|
| 515 |
+
"step": 22000
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.225,
|
| 519 |
+
"grad_norm": 0.49506717920303345,
|
| 520 |
+
"learning_rate": 3.8753e-05,
|
| 521 |
+
"loss": 6.3161,
|
| 522 |
+
"step": 22500
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.23,
|
| 526 |
+
"grad_norm": 0.6392377614974976,
|
| 527 |
+
"learning_rate": 3.8503e-05,
|
| 528 |
+
"loss": 6.3081,
|
| 529 |
+
"step": 23000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.23,
|
| 533 |
+
"eval_accuracy": 0.03635019278204852,
|
| 534 |
+
"eval_loss": 6.269931316375732,
|
| 535 |
+
"eval_runtime": 112.6212,
|
| 536 |
+
"eval_samples_per_second": 31.895,
|
| 537 |
+
"eval_steps_per_second": 1.998,
|
| 538 |
+
"step": 23000
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 0.235,
|
| 542 |
+
"grad_norm": 0.5753453969955444,
|
| 543 |
+
"learning_rate": 3.8253e-05,
|
| 544 |
+
"loss": 6.2806,
|
| 545 |
+
"step": 23500
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 0.24,
|
| 549 |
+
"grad_norm": 0.5371689796447754,
|
| 550 |
+
"learning_rate": 3.80035e-05,
|
| 551 |
+
"loss": 6.2784,
|
| 552 |
+
"step": 24000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 0.24,
|
| 556 |
+
"eval_accuracy": 0.036991620348901764,
|
| 557 |
+
"eval_loss": 6.2454352378845215,
|
| 558 |
+
"eval_runtime": 115.1822,
|
| 559 |
+
"eval_samples_per_second": 31.185,
|
| 560 |
+
"eval_steps_per_second": 1.953,
|
| 561 |
+
"step": 24000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.245,
|
| 565 |
+
"grad_norm": 0.5090768337249756,
|
| 566 |
+
"learning_rate": 3.7753500000000004e-05,
|
| 567 |
+
"loss": 6.2632,
|
| 568 |
+
"step": 24500
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 0.25,
|
| 572 |
+
"grad_norm": 0.5666926503181458,
|
| 573 |
+
"learning_rate": 3.7503500000000004e-05,
|
| 574 |
+
"loss": 6.2589,
|
| 575 |
+
"step": 25000
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.25,
|
| 579 |
+
"eval_accuracy": 0.03744921918371879,
|
| 580 |
+
"eval_loss": 6.216660499572754,
|
| 581 |
+
"eval_runtime": 112.1843,
|
| 582 |
+
"eval_samples_per_second": 32.019,
|
| 583 |
+
"eval_steps_per_second": 2.006,
|
| 584 |
+
"step": 25000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.255,
|
| 588 |
+
"grad_norm": 0.5514094233512878,
|
| 589 |
+
"learning_rate": 3.7253500000000005e-05,
|
| 590 |
+
"loss": 6.2441,
|
| 591 |
+
"step": 25500
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.26,
|
| 595 |
+
"grad_norm": 0.5683887600898743,
|
| 596 |
+
"learning_rate": 3.7004e-05,
|
| 597 |
+
"loss": 6.2371,
|
| 598 |
+
"step": 26000
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.26,
|
| 602 |
+
"eval_accuracy": 0.03695529002214109,
|
| 603 |
+
"eval_loss": 6.189018726348877,
|
| 604 |
+
"eval_runtime": 110.9014,
|
| 605 |
+
"eval_samples_per_second": 32.389,
|
| 606 |
+
"eval_steps_per_second": 2.029,
|
| 607 |
+
"step": 26000
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 0.265,
|
| 611 |
+
"grad_norm": 0.6400378346443176,
|
| 612 |
+
"learning_rate": 3.6754e-05,
|
| 613 |
+
"loss": 6.2257,
|
| 614 |
+
"step": 26500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.27,
|
| 618 |
+
"grad_norm": 0.5867771506309509,
|
| 619 |
+
"learning_rate": 3.6504e-05,
|
| 620 |
+
"loss": 6.1978,
|
| 621 |
+
"step": 27000
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 0.27,
|
| 625 |
+
"eval_accuracy": 0.0376371299749416,
|
| 626 |
+
"eval_loss": 6.165986061096191,
|
| 627 |
+
"eval_runtime": 112.4076,
|
| 628 |
+
"eval_samples_per_second": 31.955,
|
| 629 |
+
"eval_steps_per_second": 2.002,
|
| 630 |
+
"step": 27000
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 0.275,
|
| 634 |
+
"grad_norm": 0.7539021372795105,
|
| 635 |
+
"learning_rate": 3.62545e-05,
|
| 636 |
+
"loss": 6.2078,
|
| 637 |
+
"step": 27500
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 0.28,
|
| 641 |
+
"grad_norm": 0.5702797174453735,
|
| 642 |
+
"learning_rate": 3.6004500000000004e-05,
|
| 643 |
+
"loss": 6.1895,
|
| 644 |
+
"step": 28000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.28,
|
| 648 |
+
"eval_accuracy": 0.03748677412823544,
|
| 649 |
+
"eval_loss": 6.1377997398376465,
|
| 650 |
+
"eval_runtime": 111.0948,
|
| 651 |
+
"eval_samples_per_second": 32.333,
|
| 652 |
+
"eval_steps_per_second": 2.025,
|
| 653 |
+
"step": 28000
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.285,
|
| 657 |
+
"grad_norm": 0.5679642558097839,
|
| 658 |
+
"learning_rate": 3.5754500000000005e-05,
|
| 659 |
+
"loss": 6.1921,
|
| 660 |
+
"step": 28500
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.29,
|
| 664 |
+
"grad_norm": 0.6666756868362427,
|
| 665 |
+
"learning_rate": 3.55045e-05,
|
| 666 |
+
"loss": 6.1636,
|
| 667 |
+
"step": 29000
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.29,
|
| 671 |
+
"eval_accuracy": 0.03662328254163156,
|
| 672 |
+
"eval_loss": 6.121272087097168,
|
| 673 |
+
"eval_runtime": 111.2726,
|
| 674 |
+
"eval_samples_per_second": 32.281,
|
| 675 |
+
"eval_steps_per_second": 2.022,
|
| 676 |
+
"step": 29000
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.295,
|
| 680 |
+
"grad_norm": 0.5925132632255554,
|
| 681 |
+
"learning_rate": 3.52545e-05,
|
| 682 |
+
"loss": 6.1639,
|
| 683 |
+
"step": 29500
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.3,
|
| 687 |
+
"grad_norm": 0.5706653594970703,
|
| 688 |
+
"learning_rate": 3.50045e-05,
|
| 689 |
+
"loss": 6.1262,
|
| 690 |
+
"step": 30000
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.3,
|
| 694 |
+
"eval_accuracy": 0.03699611061400702,
|
| 695 |
+
"eval_loss": 6.096695899963379,
|
| 696 |
+
"eval_runtime": 112.2709,
|
| 697 |
+
"eval_samples_per_second": 31.994,
|
| 698 |
+
"eval_steps_per_second": 2.004,
|
| 699 |
+
"step": 30000
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 0.305,
|
| 703 |
+
"grad_norm": 0.6464671492576599,
|
| 704 |
+
"learning_rate": 3.47545e-05,
|
| 705 |
+
"loss": 6.1365,
|
| 706 |
+
"step": 30500
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 0.31,
|
| 710 |
+
"grad_norm": 0.5408177375793457,
|
| 711 |
+
"learning_rate": 3.4505e-05,
|
| 712 |
+
"loss": 6.1345,
|
| 713 |
+
"step": 31000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 0.31,
|
| 717 |
+
"eval_accuracy": 0.03612323029127397,
|
| 718 |
+
"eval_loss": 6.074455738067627,
|
| 719 |
+
"eval_runtime": 113.9504,
|
| 720 |
+
"eval_samples_per_second": 31.522,
|
| 721 |
+
"eval_steps_per_second": 1.975,
|
| 722 |
+
"step": 31000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.315,
|
| 726 |
+
"grad_norm": 0.5881854891777039,
|
| 727 |
+
"learning_rate": 3.4255e-05,
|
| 728 |
+
"loss": 6.1113,
|
| 729 |
+
"step": 31500
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.32,
|
| 733 |
+
"grad_norm": 0.6097517609596252,
|
| 734 |
+
"learning_rate": 3.4005000000000004e-05,
|
| 735 |
+
"loss": 6.1096,
|
| 736 |
+
"step": 32000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.32,
|
| 740 |
+
"eval_accuracy": 0.03600621126125832,
|
| 741 |
+
"eval_loss": 6.055612087249756,
|
| 742 |
+
"eval_runtime": 111.7549,
|
| 743 |
+
"eval_samples_per_second": 32.142,
|
| 744 |
+
"eval_steps_per_second": 2.013,
|
| 745 |
+
"step": 32000
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.325,
|
| 749 |
+
"grad_norm": 0.6643844246864319,
|
| 750 |
+
"learning_rate": 3.3755000000000005e-05,
|
| 751 |
+
"loss": 6.1049,
|
| 752 |
+
"step": 32500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.33,
|
| 756 |
+
"grad_norm": 0.6951597929000854,
|
| 757 |
+
"learning_rate": 3.3505000000000005e-05,
|
| 758 |
+
"loss": 6.0794,
|
| 759 |
+
"step": 33000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.33,
|
| 763 |
+
"eval_accuracy": 0.03573761176678053,
|
| 764 |
+
"eval_loss": 6.041288375854492,
|
| 765 |
+
"eval_runtime": 112.9798,
|
| 766 |
+
"eval_samples_per_second": 31.793,
|
| 767 |
+
"eval_steps_per_second": 1.992,
|
| 768 |
+
"step": 33000
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 0.335,
|
| 772 |
+
"grad_norm": 0.8083083033561707,
|
| 773 |
+
"learning_rate": 3.3255000000000006e-05,
|
| 774 |
+
"loss": 6.1019,
|
| 775 |
+
"step": 33500
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 0.34,
|
| 779 |
+
"grad_norm": 0.6118751764297485,
|
| 780 |
+
"learning_rate": 3.3005e-05,
|
| 781 |
+
"loss": 6.0643,
|
| 782 |
+
"step": 34000
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.34,
|
| 786 |
+
"eval_accuracy": 0.03625793824443153,
|
| 787 |
+
"eval_loss": 6.013612270355225,
|
| 788 |
+
"eval_runtime": 111.1363,
|
| 789 |
+
"eval_samples_per_second": 32.321,
|
| 790 |
+
"eval_steps_per_second": 2.025,
|
| 791 |
+
"step": 34000
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 0.345,
|
| 795 |
+
"grad_norm": 0.6557193398475647,
|
| 796 |
+
"learning_rate": 3.27555e-05,
|
| 797 |
+
"loss": 6.0551,
|
| 798 |
+
"step": 34500
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 0.35,
|
| 802 |
+
"grad_norm": 0.5440818667411804,
|
| 803 |
+
"learning_rate": 3.25055e-05,
|
| 804 |
+
"loss": 6.057,
|
| 805 |
+
"step": 35000
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 0.35,
|
| 809 |
+
"eval_accuracy": 0.03619099247377141,
|
| 810 |
+
"eval_loss": 5.996499538421631,
|
| 811 |
+
"eval_runtime": 114.4917,
|
| 812 |
+
"eval_samples_per_second": 31.373,
|
| 813 |
+
"eval_steps_per_second": 1.965,
|
| 814 |
+
"step": 35000
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.355,
|
| 818 |
+
"grad_norm": 0.5865129828453064,
|
| 819 |
+
"learning_rate": 3.2255499999999996e-05,
|
| 820 |
+
"loss": 6.0495,
|
| 821 |
+
"step": 35500
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.36,
|
| 825 |
+
"grad_norm": 0.598119854927063,
|
| 826 |
+
"learning_rate": 3.20055e-05,
|
| 827 |
+
"loss": 6.0337,
|
| 828 |
+
"step": 36000
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.36,
|
| 832 |
+
"eval_accuracy": 0.03535702778195055,
|
| 833 |
+
"eval_loss": 5.980613708496094,
|
| 834 |
+
"eval_runtime": 112.638,
|
| 835 |
+
"eval_samples_per_second": 31.89,
|
| 836 |
+
"eval_steps_per_second": 1.998,
|
| 837 |
+
"step": 36000
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 0.365,
|
| 841 |
+
"grad_norm": 0.5838174819946289,
|
| 842 |
+
"learning_rate": 3.1756000000000006e-05,
|
| 843 |
+
"loss": 6.0167,
|
| 844 |
+
"step": 36500
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 0.37,
|
| 848 |
+
"grad_norm": 0.7071039080619812,
|
| 849 |
+
"learning_rate": 3.1506e-05,
|
| 850 |
+
"loss": 6.0217,
|
| 851 |
+
"step": 37000
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 0.37,
|
| 855 |
+
"eval_accuracy": 0.03629073078656382,
|
| 856 |
+
"eval_loss": 5.958355903625488,
|
| 857 |
+
"eval_runtime": 112.2627,
|
| 858 |
+
"eval_samples_per_second": 31.996,
|
| 859 |
+
"eval_steps_per_second": 2.004,
|
| 860 |
+
"step": 37000
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 0.375,
|
| 864 |
+
"grad_norm": 0.606131374835968,
|
| 865 |
+
"learning_rate": 3.1256e-05,
|
| 866 |
+
"loss": 6.0086,
|
| 867 |
+
"step": 37500
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 0.38,
|
| 871 |
+
"grad_norm": 0.6173387169837952,
|
| 872 |
+
"learning_rate": 3.1006e-05,
|
| 873 |
+
"loss": 6.0045,
|
| 874 |
+
"step": 38000
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 0.38,
|
| 878 |
+
"eval_accuracy": 0.035918719126025685,
|
| 879 |
+
"eval_loss": 5.952617168426514,
|
| 880 |
+
"eval_runtime": 116.0654,
|
| 881 |
+
"eval_samples_per_second": 30.948,
|
| 882 |
+
"eval_steps_per_second": 1.939,
|
| 883 |
+
"step": 38000
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 0.385,
|
| 887 |
+
"grad_norm": 0.6699332594871521,
|
| 888 |
+
"learning_rate": 3.0756e-05,
|
| 889 |
+
"loss": 5.9996,
|
| 890 |
+
"step": 38500
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.39,
|
| 894 |
+
"grad_norm": 0.6398167610168457,
|
| 895 |
+
"learning_rate": 3.0506000000000002e-05,
|
| 896 |
+
"loss": 5.9896,
|
| 897 |
+
"step": 39000
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.39,
|
| 901 |
+
"eval_accuracy": 0.0355204462180538,
|
| 902 |
+
"eval_loss": 5.9288482666015625,
|
| 903 |
+
"eval_runtime": 110.5071,
|
| 904 |
+
"eval_samples_per_second": 32.505,
|
| 905 |
+
"eval_steps_per_second": 2.036,
|
| 906 |
+
"step": 39000
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.395,
|
| 910 |
+
"grad_norm": 0.7196072936058044,
|
| 911 |
+
"learning_rate": 3.0256e-05,
|
| 912 |
+
"loss": 5.9879,
|
| 913 |
+
"step": 39500
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.4,
|
| 917 |
+
"grad_norm": 0.6065074801445007,
|
| 918 |
+
"learning_rate": 3.0006e-05,
|
| 919 |
+
"loss": 5.9711,
|
| 920 |
+
"step": 40000
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.4,
|
| 924 |
+
"eval_accuracy": 0.03516381031378517,
|
| 925 |
+
"eval_loss": 5.915222644805908,
|
| 926 |
+
"eval_runtime": 110.7645,
|
| 927 |
+
"eval_samples_per_second": 32.429,
|
| 928 |
+
"eval_steps_per_second": 2.031,
|
| 929 |
+
"step": 40000
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.405,
|
| 933 |
+
"grad_norm": 0.6913782954216003,
|
| 934 |
+
"learning_rate": 2.9757e-05,
|
| 935 |
+
"loss": 5.9671,
|
| 936 |
+
"step": 40500
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 0.41,
|
| 940 |
+
"grad_norm": 0.6546822190284729,
|
| 941 |
+
"learning_rate": 2.9507e-05,
|
| 942 |
+
"loss": 5.9629,
|
| 943 |
+
"step": 41000
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 0.41,
|
| 947 |
+
"eval_accuracy": 0.03489303372107453,
|
| 948 |
+
"eval_loss": 5.896161079406738,
|
| 949 |
+
"eval_runtime": 111.3465,
|
| 950 |
+
"eval_samples_per_second": 32.26,
|
| 951 |
+
"eval_steps_per_second": 2.021,
|
| 952 |
+
"step": 41000
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 0.415,
|
| 956 |
+
"grad_norm": 0.6649766564369202,
|
| 957 |
+
"learning_rate": 2.9257e-05,
|
| 958 |
+
"loss": 5.9631,
|
| 959 |
+
"step": 41500
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 0.42,
|
| 963 |
+
"grad_norm": 0.7166668176651001,
|
| 964 |
+
"learning_rate": 2.9007000000000002e-05,
|
| 965 |
+
"loss": 5.9465,
|
| 966 |
+
"step": 42000
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 0.42,
|
| 970 |
+
"eval_accuracy": 0.035887695476207584,
|
| 971 |
+
"eval_loss": 5.882122993469238,
|
| 972 |
+
"eval_runtime": 110.5783,
|
| 973 |
+
"eval_samples_per_second": 32.484,
|
| 974 |
+
"eval_steps_per_second": 2.035,
|
| 975 |
+
"step": 42000
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.425,
|
| 979 |
+
"grad_norm": 0.6148796081542969,
|
| 980 |
+
"learning_rate": 2.8757e-05,
|
| 981 |
+
"loss": 5.9383,
|
| 982 |
+
"step": 42500
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.43,
|
| 986 |
+
"grad_norm": 0.712359607219696,
|
| 987 |
+
"learning_rate": 2.8507e-05,
|
| 988 |
+
"loss": 5.9463,
|
| 989 |
+
"step": 43000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.43,
|
| 993 |
+
"eval_accuracy": 0.034527145149316284,
|
| 994 |
+
"eval_loss": 5.869154930114746,
|
| 995 |
+
"eval_runtime": 113.9173,
|
| 996 |
+
"eval_samples_per_second": 31.532,
|
| 997 |
+
"eval_steps_per_second": 1.975,
|
| 998 |
+
"step": 43000
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 0.435,
|
| 1002 |
+
"grad_norm": 0.6819197535514832,
|
| 1003 |
+
"learning_rate": 2.8257e-05,
|
| 1004 |
+
"loss": 5.9363,
|
| 1005 |
+
"step": 43500
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 0.44,
|
| 1009 |
+
"grad_norm": 0.7428186535835266,
|
| 1010 |
+
"learning_rate": 2.8007e-05,
|
| 1011 |
+
"loss": 5.9317,
|
| 1012 |
+
"step": 44000
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 0.44,
|
| 1016 |
+
"eval_accuracy": 0.034260450615792234,
|
| 1017 |
+
"eval_loss": 5.869858741760254,
|
| 1018 |
+
"eval_runtime": 111.9079,
|
| 1019 |
+
"eval_samples_per_second": 32.098,
|
| 1020 |
+
"eval_steps_per_second": 2.011,
|
| 1021 |
+
"step": 44000
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 0.445,
|
| 1025 |
+
"grad_norm": 0.6326854825019836,
|
| 1026 |
+
"learning_rate": 2.7757500000000003e-05,
|
| 1027 |
+
"loss": 5.9114,
|
| 1028 |
+
"step": 44500
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 1.00345,
|
| 1032 |
+
"grad_norm": 0.6722562313079834,
|
| 1033 |
+
"learning_rate": 2.7507500000000004e-05,
|
| 1034 |
+
"loss": 5.9097,
|
| 1035 |
+
"step": 45000
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 1.00345,
|
| 1039 |
+
"eval_accuracy": 0.034564291887914274,
|
| 1040 |
+
"eval_loss": 5.8482561111450195,
|
| 1041 |
+
"eval_runtime": 111.3739,
|
| 1042 |
+
"eval_samples_per_second": 32.252,
|
| 1043 |
+
"eval_steps_per_second": 2.02,
|
| 1044 |
+
"step": 45000
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 1.00845,
|
| 1048 |
+
"grad_norm": 0.6160574555397034,
|
| 1049 |
+
"learning_rate": 2.72575e-05,
|
| 1050 |
+
"loss": 5.8948,
|
| 1051 |
+
"step": 45500
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 1.01345,
|
| 1055 |
+
"grad_norm": 0.6146988868713379,
|
| 1056 |
+
"learning_rate": 2.7007500000000002e-05,
|
| 1057 |
+
"loss": 5.9107,
|
| 1058 |
+
"step": 46000
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 1.01345,
|
| 1062 |
+
"eval_accuracy": 0.03477846392657083,
|
| 1063 |
+
"eval_loss": 5.8351731300354,
|
| 1064 |
+
"eval_runtime": 112.8785,
|
| 1065 |
+
"eval_samples_per_second": 31.822,
|
| 1066 |
+
"eval_steps_per_second": 1.993,
|
| 1067 |
+
"step": 46000
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.01845,
|
| 1071 |
+
"grad_norm": 0.6342368721961975,
|
| 1072 |
+
"learning_rate": 2.6758e-05,
|
| 1073 |
+
"loss": 5.893,
|
| 1074 |
+
"step": 46500
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.02345,
|
| 1078 |
+
"grad_norm": 0.7240073680877686,
|
| 1079 |
+
"learning_rate": 2.6508e-05,
|
| 1080 |
+
"loss": 5.8838,
|
| 1081 |
+
"step": 47000
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.02345,
|
| 1085 |
+
"eval_accuracy": 0.034256504625245196,
|
| 1086 |
+
"eval_loss": 5.818812847137451,
|
| 1087 |
+
"eval_runtime": 112.1297,
|
| 1088 |
+
"eval_samples_per_second": 32.034,
|
| 1089 |
+
"eval_steps_per_second": 2.007,
|
| 1090 |
+
"step": 47000
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 1.02845,
|
| 1094 |
+
"grad_norm": 0.6988590955734253,
|
| 1095 |
+
"learning_rate": 2.6257999999999998e-05,
|
| 1096 |
+
"loss": 5.8907,
|
| 1097 |
+
"step": 47500
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 1.03345,
|
| 1101 |
+
"grad_norm": 0.6514461636543274,
|
| 1102 |
+
"learning_rate": 2.6008e-05,
|
| 1103 |
+
"loss": 5.887,
|
| 1104 |
+
"step": 48000
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 1.03345,
|
| 1108 |
+
"eval_accuracy": 0.03395511258863511,
|
| 1109 |
+
"eval_loss": 5.808635234832764,
|
| 1110 |
+
"eval_runtime": 112.3746,
|
| 1111 |
+
"eval_samples_per_second": 31.965,
|
| 1112 |
+
"eval_steps_per_second": 2.002,
|
| 1113 |
+
"step": 48000
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 1.03845,
|
| 1117 |
+
"grad_norm": 0.6310043931007385,
|
| 1118 |
+
"learning_rate": 2.5758e-05,
|
| 1119 |
+
"loss": 5.873,
|
| 1120 |
+
"step": 48500
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 1.04345,
|
| 1124 |
+
"grad_norm": 0.7601247429847717,
|
| 1125 |
+
"learning_rate": 2.5507999999999997e-05,
|
| 1126 |
+
"loss": 5.8563,
|
| 1127 |
+
"step": 49000
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 1.04345,
|
| 1131 |
+
"eval_accuracy": 0.03384244775508516,
|
| 1132 |
+
"eval_loss": 5.797094821929932,
|
| 1133 |
+
"eval_runtime": 112.1133,
|
| 1134 |
+
"eval_samples_per_second": 32.039,
|
| 1135 |
+
"eval_steps_per_second": 2.007,
|
| 1136 |
+
"step": 49000
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 1.04845,
|
| 1140 |
+
"grad_norm": 0.6604752540588379,
|
| 1141 |
+
"learning_rate": 2.5258500000000002e-05,
|
| 1142 |
+
"loss": 5.8583,
|
| 1143 |
+
"step": 49500
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 1.05345,
|
| 1147 |
+
"grad_norm": 0.681904673576355,
|
| 1148 |
+
"learning_rate": 2.5008500000000003e-05,
|
| 1149 |
+
"loss": 5.8576,
|
| 1150 |
+
"step": 50000
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 1.05345,
|
| 1154 |
+
"eval_accuracy": 0.03392898740984092,
|
| 1155 |
+
"eval_loss": 5.796751022338867,
|
| 1156 |
+
"eval_runtime": 113.244,
|
| 1157 |
+
"eval_samples_per_second": 31.719,
|
| 1158 |
+
"eval_steps_per_second": 1.987,
|
| 1159 |
+
"step": 50000
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 1.0584500000000001,
|
| 1163 |
+
"grad_norm": 0.6337283849716187,
|
| 1164 |
+
"learning_rate": 2.47585e-05,
|
| 1165 |
+
"loss": 5.8495,
|
| 1166 |
+
"step": 50500
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 1.06345,
|
| 1170 |
+
"grad_norm": 0.6972376108169556,
|
| 1171 |
+
"learning_rate": 2.45085e-05,
|
| 1172 |
+
"loss": 5.8567,
|
| 1173 |
+
"step": 51000
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 1.06345,
|
| 1177 |
+
"eval_accuracy": 0.034259225998036255,
|
| 1178 |
+
"eval_loss": 5.779718399047852,
|
| 1179 |
+
"eval_runtime": 110.7791,
|
| 1180 |
+
"eval_samples_per_second": 32.425,
|
| 1181 |
+
"eval_steps_per_second": 2.031,
|
| 1182 |
+
"step": 51000
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 1.06845,
|
| 1186 |
+
"grad_norm": 0.6530519127845764,
|
| 1187 |
+
"learning_rate": 2.42585e-05,
|
| 1188 |
+
"loss": 5.8519,
|
| 1189 |
+
"step": 51500
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 1.07345,
|
| 1193 |
+
"grad_norm": 0.6328603625297546,
|
| 1194 |
+
"learning_rate": 2.4009e-05,
|
| 1195 |
+
"loss": 5.841,
|
| 1196 |
+
"step": 52000
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 1.07345,
|
| 1200 |
+
"eval_accuracy": 0.033661476464479555,
|
| 1201 |
+
"eval_loss": 5.767716407775879,
|
| 1202 |
+
"eval_runtime": 114.652,
|
| 1203 |
+
"eval_samples_per_second": 31.33,
|
| 1204 |
+
"eval_steps_per_second": 1.962,
|
| 1205 |
+
"step": 52000
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 1.07845,
|
| 1209 |
+
"grad_norm": 0.6603217124938965,
|
| 1210 |
+
"learning_rate": 2.3759e-05,
|
| 1211 |
+
"loss": 5.8284,
|
| 1212 |
+
"step": 52500
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 1.08345,
|
| 1216 |
+
"grad_norm": 0.6852928400039673,
|
| 1217 |
+
"learning_rate": 2.3509e-05,
|
| 1218 |
+
"loss": 5.8192,
|
| 1219 |
+
"step": 53000
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 1.08345,
|
| 1223 |
+
"eval_accuracy": 0.03317965741182208,
|
| 1224 |
+
"eval_loss": 5.76131534576416,
|
| 1225 |
+
"eval_runtime": 111.561,
|
| 1226 |
+
"eval_samples_per_second": 32.198,
|
| 1227 |
+
"eval_steps_per_second": 2.017,
|
| 1228 |
+
"step": 53000
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.08845,
|
| 1232 |
+
"grad_norm": 0.6923466324806213,
|
| 1233 |
+
"learning_rate": 2.3259e-05,
|
| 1234 |
+
"loss": 5.8147,
|
| 1235 |
+
"step": 53500
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.09345,
|
| 1239 |
+
"grad_norm": 0.7163240313529968,
|
| 1240 |
+
"learning_rate": 2.3009e-05,
|
| 1241 |
+
"loss": 5.8214,
|
| 1242 |
+
"step": 54000
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.09345,
|
| 1246 |
+
"eval_accuracy": 0.03383822962725901,
|
| 1247 |
+
"eval_loss": 5.748566150665283,
|
| 1248 |
+
"eval_runtime": 113.5245,
|
| 1249 |
+
"eval_samples_per_second": 31.641,
|
| 1250 |
+
"eval_steps_per_second": 1.982,
|
| 1251 |
+
"step": 54000
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 1.09845,
|
| 1255 |
+
"grad_norm": 0.7053197026252747,
|
| 1256 |
+
"learning_rate": 2.27595e-05,
|
| 1257 |
+
"loss": 5.8176,
|
| 1258 |
+
"step": 54500
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 1.10345,
|
| 1262 |
+
"grad_norm": 0.7318024039268494,
|
| 1263 |
+
"learning_rate": 2.25095e-05,
|
| 1264 |
+
"loss": 5.8166,
|
| 1265 |
+
"step": 55000
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 1.10345,
|
| 1269 |
+
"eval_accuracy": 0.033826935930176105,
|
| 1270 |
+
"eval_loss": 5.74090576171875,
|
| 1271 |
+
"eval_runtime": 112.6548,
|
| 1272 |
+
"eval_samples_per_second": 31.885,
|
| 1273 |
+
"eval_steps_per_second": 1.997,
|
| 1274 |
+
"step": 55000
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 1.10845,
|
| 1278 |
+
"grad_norm": 0.727443516254425,
|
| 1279 |
+
"learning_rate": 2.22595e-05,
|
| 1280 |
+
"loss": 5.8054,
|
| 1281 |
+
"step": 55500
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 1.11345,
|
| 1285 |
+
"grad_norm": 0.6675511598587036,
|
| 1286 |
+
"learning_rate": 2.2009500000000003e-05,
|
| 1287 |
+
"loss": 5.806,
|
| 1288 |
+
"step": 56000
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 1.11345,
|
| 1292 |
+
"eval_accuracy": 0.03327177588079952,
|
| 1293 |
+
"eval_loss": 5.734244346618652,
|
| 1294 |
+
"eval_runtime": 114.861,
|
| 1295 |
+
"eval_samples_per_second": 31.273,
|
| 1296 |
+
"eval_steps_per_second": 1.959,
|
| 1297 |
+
"step": 56000
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 1.11845,
|
| 1301 |
+
"grad_norm": 0.7548192143440247,
|
| 1302 |
+
"learning_rate": 2.17595e-05,
|
| 1303 |
+
"loss": 5.7987,
|
| 1304 |
+
"step": 56500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 1.12345,
|
| 1308 |
+
"grad_norm": 0.6868897676467896,
|
| 1309 |
+
"learning_rate": 2.15095e-05,
|
| 1310 |
+
"loss": 5.7961,
|
| 1311 |
+
"step": 57000
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 1.12345,
|
| 1315 |
+
"eval_accuracy": 0.033481321585711266,
|
| 1316 |
+
"eval_loss": 5.723623752593994,
|
| 1317 |
+
"eval_runtime": 113.4262,
|
| 1318 |
+
"eval_samples_per_second": 31.668,
|
| 1319 |
+
"eval_steps_per_second": 1.984,
|
| 1320 |
+
"step": 57000
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 1.12845,
|
| 1324 |
+
"grad_norm": 0.6014879941940308,
|
| 1325 |
+
"learning_rate": 2.1259500000000002e-05,
|
| 1326 |
+
"loss": 5.7892,
|
| 1327 |
+
"step": 57500
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 1.13345,
|
| 1331 |
+
"grad_norm": 0.6603185534477234,
|
| 1332 |
+
"learning_rate": 2.101e-05,
|
| 1333 |
+
"loss": 5.7847,
|
| 1334 |
+
"step": 58000
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 1.13345,
|
| 1338 |
+
"eval_accuracy": 0.033258985428681526,
|
| 1339 |
+
"eval_loss": 5.716407775878906,
|
| 1340 |
+
"eval_runtime": 112.5491,
|
| 1341 |
+
"eval_samples_per_second": 31.915,
|
| 1342 |
+
"eval_steps_per_second": 1.999,
|
| 1343 |
+
"step": 58000
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 1.13845,
|
| 1347 |
+
"grad_norm": 0.7514461278915405,
|
| 1348 |
+
"learning_rate": 2.076e-05,
|
| 1349 |
+
"loss": 5.7785,
|
| 1350 |
+
"step": 58500
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 1.14345,
|
| 1354 |
+
"grad_norm": 0.9591528177261353,
|
| 1355 |
+
"learning_rate": 2.0510000000000002e-05,
|
| 1356 |
+
"loss": 5.787,
|
| 1357 |
+
"step": 59000
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"epoch": 1.14345,
|
| 1361 |
+
"eval_accuracy": 0.03300685023958966,
|
| 1362 |
+
"eval_loss": 5.7095818519592285,
|
| 1363 |
+
"eval_runtime": 112.4202,
|
| 1364 |
+
"eval_samples_per_second": 31.952,
|
| 1365 |
+
"eval_steps_per_second": 2.001,
|
| 1366 |
+
"step": 59000
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 1.14845,
|
| 1370 |
+
"grad_norm": 0.6442362666130066,
|
| 1371 |
+
"learning_rate": 2.0260000000000003e-05,
|
| 1372 |
+
"loss": 5.7765,
|
| 1373 |
+
"step": 59500
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 1.15345,
|
| 1377 |
+
"grad_norm": 0.7159613370895386,
|
| 1378 |
+
"learning_rate": 2.001e-05,
|
| 1379 |
+
"loss": 5.7711,
|
| 1380 |
+
"step": 60000
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 1.15345,
|
| 1384 |
+
"eval_accuracy": 0.03279621598556148,
|
| 1385 |
+
"eval_loss": 5.703481674194336,
|
| 1386 |
+
"eval_runtime": 112.2964,
|
| 1387 |
+
"eval_samples_per_second": 31.987,
|
| 1388 |
+
"eval_steps_per_second": 2.004,
|
| 1389 |
+
"step": 60000
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.15845,
|
| 1393 |
+
"grad_norm": 0.6875385642051697,
|
| 1394 |
+
"learning_rate": 1.976e-05,
|
| 1395 |
+
"loss": 5.7593,
|
| 1396 |
+
"step": 60500
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.16345,
|
| 1400 |
+
"grad_norm": 0.6751101016998291,
|
| 1401 |
+
"learning_rate": 1.951e-05,
|
| 1402 |
+
"loss": 5.7699,
|
| 1403 |
+
"step": 61000
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.16345,
|
| 1407 |
+
"eval_accuracy": 0.03306223017588777,
|
| 1408 |
+
"eval_loss": 5.6887898445129395,
|
| 1409 |
+
"eval_runtime": 113.7541,
|
| 1410 |
+
"eval_samples_per_second": 31.577,
|
| 1411 |
+
"eval_steps_per_second": 1.978,
|
| 1412 |
+
"step": 61000
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 1.16845,
|
| 1416 |
+
"grad_norm": 0.7514244914054871,
|
| 1417 |
+
"learning_rate": 1.92605e-05,
|
| 1418 |
+
"loss": 5.7672,
|
| 1419 |
+
"step": 61500
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 1.1734499999999999,
|
| 1423 |
+
"grad_norm": 0.697762131690979,
|
| 1424 |
+
"learning_rate": 1.90105e-05,
|
| 1425 |
+
"loss": 5.763,
|
| 1426 |
+
"step": 62000
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 1.1734499999999999,
|
| 1430 |
+
"eval_accuracy": 0.03338308002795394,
|
| 1431 |
+
"eval_loss": 5.687499046325684,
|
| 1432 |
+
"eval_runtime": 111.5862,
|
| 1433 |
+
"eval_samples_per_second": 32.19,
|
| 1434 |
+
"eval_steps_per_second": 2.016,
|
| 1435 |
+
"step": 62000
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 1.17845,
|
| 1439 |
+
"grad_norm": 0.6761494874954224,
|
| 1440 |
+
"learning_rate": 1.87605e-05,
|
| 1441 |
+
"loss": 5.7669,
|
| 1442 |
+
"step": 62500
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 1.1834500000000001,
|
| 1446 |
+
"grad_norm": 0.6467716693878174,
|
| 1447 |
+
"learning_rate": 1.8510500000000002e-05,
|
| 1448 |
+
"loss": 5.7434,
|
| 1449 |
+
"step": 63000
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 1.1834500000000001,
|
| 1453 |
+
"eval_accuracy": 0.03302726053552262,
|
| 1454 |
+
"eval_loss": 5.680927276611328,
|
| 1455 |
+
"eval_runtime": 112.7327,
|
| 1456 |
+
"eval_samples_per_second": 31.863,
|
| 1457 |
+
"eval_steps_per_second": 1.996,
|
| 1458 |
+
"step": 63000
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 1.18845,
|
| 1462 |
+
"grad_norm": 0.7104445695877075,
|
| 1463 |
+
"learning_rate": 1.82605e-05,
|
| 1464 |
+
"loss": 5.7506,
|
| 1465 |
+
"step": 63500
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 1.19345,
|
| 1469 |
+
"grad_norm": 0.6914287805557251,
|
| 1470 |
+
"learning_rate": 1.8011e-05,
|
| 1471 |
+
"loss": 5.7477,
|
| 1472 |
+
"step": 64000
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 1.19345,
|
| 1476 |
+
"eval_accuracy": 0.032914323564693565,
|
| 1477 |
+
"eval_loss": 5.668553829193115,
|
| 1478 |
+
"eval_runtime": 113.1627,
|
| 1479 |
+
"eval_samples_per_second": 31.742,
|
| 1480 |
+
"eval_steps_per_second": 1.988,
|
| 1481 |
+
"step": 64000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 1.19845,
|
| 1485 |
+
"grad_norm": 0.7188768982887268,
|
| 1486 |
+
"learning_rate": 1.7760999999999998e-05,
|
| 1487 |
+
"loss": 5.7329,
|
| 1488 |
+
"step": 64500
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 1.20345,
|
| 1492 |
+
"grad_norm": 0.6479863524436951,
|
| 1493 |
+
"learning_rate": 1.7511e-05,
|
| 1494 |
+
"loss": 5.7409,
|
| 1495 |
+
"step": 65000
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 1.20345,
|
| 1499 |
+
"eval_accuracy": 0.03304100346811748,
|
| 1500 |
+
"eval_loss": 5.662350654602051,
|
| 1501 |
+
"eval_runtime": 110.8399,
|
| 1502 |
+
"eval_samples_per_second": 32.407,
|
| 1503 |
+
"eval_steps_per_second": 2.03,
|
| 1504 |
+
"step": 65000
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 1.20845,
|
| 1508 |
+
"grad_norm": 0.678534209728241,
|
| 1509 |
+
"learning_rate": 1.7261000000000003e-05,
|
| 1510 |
+
"loss": 5.7384,
|
| 1511 |
+
"step": 65500
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 1.21345,
|
| 1515 |
+
"grad_norm": 0.8188093900680542,
|
| 1516 |
+
"learning_rate": 1.7011e-05,
|
| 1517 |
+
"loss": 5.737,
|
| 1518 |
+
"step": 66000
|
| 1519 |
+
},
|
| 1520 |
+
{
|
| 1521 |
+
"epoch": 1.21345,
|
| 1522 |
+
"eval_accuracy": 0.03385346931488896,
|
| 1523 |
+
"eval_loss": 5.675750732421875,
|
| 1524 |
+
"eval_runtime": 111.9995,
|
| 1525 |
+
"eval_samples_per_second": 32.072,
|
| 1526 |
+
"eval_steps_per_second": 2.009,
|
| 1527 |
+
"step": 66000
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 1.21845,
|
| 1531 |
+
"grad_norm": 0.7928422689437866,
|
| 1532 |
+
"learning_rate": 1.6761e-05,
|
| 1533 |
+
"loss": 5.7299,
|
| 1534 |
+
"step": 66500
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 1.22345,
|
| 1538 |
+
"grad_norm": 0.7139099836349487,
|
| 1539 |
+
"learning_rate": 1.6511e-05,
|
| 1540 |
+
"loss": 5.729,
|
| 1541 |
+
"step": 67000
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"epoch": 1.22345,
|
| 1545 |
+
"eval_accuracy": 0.032598372183651296,
|
| 1546 |
+
"eval_loss": 5.654570579528809,
|
| 1547 |
+
"eval_runtime": 111.7834,
|
| 1548 |
+
"eval_samples_per_second": 32.134,
|
| 1549 |
+
"eval_steps_per_second": 2.013,
|
| 1550 |
+
"step": 67000
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.22845,
|
| 1554 |
+
"grad_norm": 0.6944179534912109,
|
| 1555 |
+
"learning_rate": 1.6261000000000002e-05,
|
| 1556 |
+
"loss": 5.7183,
|
| 1557 |
+
"step": 67500
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.23345,
|
| 1561 |
+
"grad_norm": 0.7094106078147888,
|
| 1562 |
+
"learning_rate": 1.60115e-05,
|
| 1563 |
+
"loss": 5.7232,
|
| 1564 |
+
"step": 68000
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.23345,
|
| 1568 |
+
"eval_accuracy": 0.032930243595521276,
|
| 1569 |
+
"eval_loss": 5.646746635437012,
|
| 1570 |
+
"eval_runtime": 117.6522,
|
| 1571 |
+
"eval_samples_per_second": 30.531,
|
| 1572 |
+
"eval_steps_per_second": 1.912,
|
| 1573 |
+
"step": 68000
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 1.23845,
|
| 1577 |
+
"grad_norm": 0.6927877068519592,
|
| 1578 |
+
"learning_rate": 1.57615e-05,
|
| 1579 |
+
"loss": 5.7162,
|
| 1580 |
+
"step": 68500
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 1.24345,
|
| 1584 |
+
"grad_norm": 0.7557797431945801,
|
| 1585 |
+
"learning_rate": 1.5511500000000002e-05,
|
| 1586 |
+
"loss": 5.7127,
|
| 1587 |
+
"step": 69000
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 1.24345,
|
| 1591 |
+
"eval_accuracy": 0.032887245905422496,
|
| 1592 |
+
"eval_loss": 5.644942283630371,
|
| 1593 |
+
"eval_runtime": 111.9326,
|
| 1594 |
+
"eval_samples_per_second": 32.091,
|
| 1595 |
+
"eval_steps_per_second": 2.01,
|
| 1596 |
+
"step": 69000
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 1.24845,
|
| 1600 |
+
"grad_norm": 0.673581063747406,
|
| 1601 |
+
"learning_rate": 1.52615e-05,
|
| 1602 |
+
"loss": 5.7094,
|
| 1603 |
+
"step": 69500
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 1.25345,
|
| 1607 |
+
"grad_norm": 0.6678490042686462,
|
| 1608 |
+
"learning_rate": 1.50115e-05,
|
| 1609 |
+
"loss": 5.7187,
|
| 1610 |
+
"step": 70000
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 1.25345,
|
| 1614 |
+
"eval_accuracy": 0.03288316384623591,
|
| 1615 |
+
"eval_loss": 5.635218143463135,
|
| 1616 |
+
"eval_runtime": 116.4191,
|
| 1617 |
+
"eval_samples_per_second": 30.854,
|
| 1618 |
+
"eval_steps_per_second": 1.933,
|
| 1619 |
+
"step": 70000
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 1.25845,
|
| 1623 |
+
"grad_norm": 0.7218269109725952,
|
| 1624 |
+
"learning_rate": 1.4761500000000001e-05,
|
| 1625 |
+
"loss": 5.7138,
|
| 1626 |
+
"step": 70500
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 1.26345,
|
| 1630 |
+
"grad_norm": 0.7058696150779724,
|
| 1631 |
+
"learning_rate": 1.4512000000000001e-05,
|
| 1632 |
+
"loss": 5.717,
|
| 1633 |
+
"step": 71000
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 1.26345,
|
| 1637 |
+
"eval_accuracy": 0.03260463134107074,
|
| 1638 |
+
"eval_loss": 5.626367568969727,
|
| 1639 |
+
"eval_runtime": 111.4646,
|
| 1640 |
+
"eval_samples_per_second": 32.225,
|
| 1641 |
+
"eval_steps_per_second": 2.019,
|
| 1642 |
+
"step": 71000
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 1.26845,
|
| 1646 |
+
"grad_norm": 0.7154064774513245,
|
| 1647 |
+
"learning_rate": 1.4262e-05,
|
| 1648 |
+
"loss": 5.6955,
|
| 1649 |
+
"step": 71500
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 1.27345,
|
| 1653 |
+
"grad_norm": 0.6907570362091064,
|
| 1654 |
+
"learning_rate": 1.4012e-05,
|
| 1655 |
+
"loss": 5.714,
|
| 1656 |
+
"step": 72000
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 1.27345,
|
| 1660 |
+
"eval_accuracy": 0.03297637086432977,
|
| 1661 |
+
"eval_loss": 5.6219401359558105,
|
| 1662 |
+
"eval_runtime": 113.1231,
|
| 1663 |
+
"eval_samples_per_second": 31.753,
|
| 1664 |
+
"eval_steps_per_second": 1.989,
|
| 1665 |
+
"step": 72000
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 1.2784499999999999,
|
| 1669 |
+
"grad_norm": 0.708003580570221,
|
| 1670 |
+
"learning_rate": 1.3762e-05,
|
| 1671 |
+
"loss": 5.6993,
|
| 1672 |
+
"step": 72500
|
| 1673 |
+
},
|
| 1674 |
+
{
|
| 1675 |
+
"epoch": 1.28345,
|
| 1676 |
+
"grad_norm": 0.8350797295570374,
|
| 1677 |
+
"learning_rate": 1.35125e-05,
|
| 1678 |
+
"loss": 5.7079,
|
| 1679 |
+
"step": 73000
|
| 1680 |
+
},
|
| 1681 |
+
{
|
| 1682 |
+
"epoch": 1.28345,
|
| 1683 |
+
"eval_accuracy": 0.032977323344806644,
|
| 1684 |
+
"eval_loss": 5.616916656494141,
|
| 1685 |
+
"eval_runtime": 112.8971,
|
| 1686 |
+
"eval_samples_per_second": 31.817,
|
| 1687 |
+
"eval_steps_per_second": 1.993,
|
| 1688 |
+
"step": 73000
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 1.28845,
|
| 1692 |
+
"grad_norm": 0.6496825218200684,
|
| 1693 |
+
"learning_rate": 1.32625e-05,
|
| 1694 |
+
"loss": 5.7047,
|
| 1695 |
+
"step": 73500
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 1.29345,
|
| 1699 |
+
"grad_norm": 0.7288230657577515,
|
| 1700 |
+
"learning_rate": 1.30125e-05,
|
| 1701 |
+
"loss": 5.7034,
|
| 1702 |
+
"step": 74000
|
| 1703 |
+
},
|
| 1704 |
+
{
|
| 1705 |
+
"epoch": 1.29345,
|
| 1706 |
+
"eval_accuracy": 0.032624225225166385,
|
| 1707 |
+
"eval_loss": 5.613090991973877,
|
| 1708 |
+
"eval_runtime": 112.7724,
|
| 1709 |
+
"eval_samples_per_second": 31.852,
|
| 1710 |
+
"eval_steps_per_second": 1.995,
|
| 1711 |
+
"step": 74000
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.2984499999999999,
|
| 1715 |
+
"grad_norm": 0.6955094337463379,
|
| 1716 |
+
"learning_rate": 1.27625e-05,
|
| 1717 |
+
"loss": 5.6884,
|
| 1718 |
+
"step": 74500
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.30345,
|
| 1722 |
+
"grad_norm": 0.6996705532073975,
|
| 1723 |
+
"learning_rate": 1.25125e-05,
|
| 1724 |
+
"loss": 5.6768,
|
| 1725 |
+
"step": 75000
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.30345,
|
| 1729 |
+
"eval_accuracy": 0.03247931212404235,
|
| 1730 |
+
"eval_loss": 5.61249303817749,
|
| 1731 |
+
"eval_runtime": 113.0871,
|
| 1732 |
+
"eval_samples_per_second": 31.763,
|
| 1733 |
+
"eval_steps_per_second": 1.99,
|
| 1734 |
+
"step": 75000
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 1.3084500000000001,
|
| 1738 |
+
"grad_norm": 0.7164750695228577,
|
| 1739 |
+
"learning_rate": 1.22625e-05,
|
| 1740 |
+
"loss": 5.685,
|
| 1741 |
+
"step": 75500
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"epoch": 1.31345,
|
| 1745 |
+
"grad_norm": 0.6893213391304016,
|
| 1746 |
+
"learning_rate": 1.20125e-05,
|
| 1747 |
+
"loss": 5.6955,
|
| 1748 |
+
"step": 76000
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 1.31345,
|
| 1752 |
+
"eval_accuracy": 0.0328188033797273,
|
| 1753 |
+
"eval_loss": 5.6074957847595215,
|
| 1754 |
+
"eval_runtime": 114.2316,
|
| 1755 |
+
"eval_samples_per_second": 31.445,
|
| 1756 |
+
"eval_steps_per_second": 1.97,
|
| 1757 |
+
"step": 76000
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 1.31845,
|
| 1761 |
+
"grad_norm": 0.7451682090759277,
|
| 1762 |
+
"learning_rate": 1.17625e-05,
|
| 1763 |
+
"loss": 5.6853,
|
| 1764 |
+
"step": 76500
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 1.32345,
|
| 1768 |
+
"grad_norm": 0.7856729626655579,
|
| 1769 |
+
"learning_rate": 1.15125e-05,
|
| 1770 |
+
"loss": 5.6947,
|
| 1771 |
+
"step": 77000
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 1.32345,
|
| 1775 |
+
"eval_accuracy": 0.032477407163088605,
|
| 1776 |
+
"eval_loss": 5.601708889007568,
|
| 1777 |
+
"eval_runtime": 111.6415,
|
| 1778 |
+
"eval_samples_per_second": 32.174,
|
| 1779 |
+
"eval_steps_per_second": 2.015,
|
| 1780 |
+
"step": 77000
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 1.3284500000000001,
|
| 1784 |
+
"grad_norm": 0.7268499135971069,
|
| 1785 |
+
"learning_rate": 1.1262500000000001e-05,
|
| 1786 |
+
"loss": 5.6676,
|
| 1787 |
+
"step": 77500
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 1.33345,
|
| 1791 |
+
"grad_norm": 0.6918110847473145,
|
| 1792 |
+
"learning_rate": 1.1013000000000001e-05,
|
| 1793 |
+
"loss": 5.7056,
|
| 1794 |
+
"step": 78000
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 1.33345,
|
| 1798 |
+
"eval_accuracy": 0.032296980147041215,
|
| 1799 |
+
"eval_loss": 5.595623970031738,
|
| 1800 |
+
"eval_runtime": 110.9779,
|
| 1801 |
+
"eval_samples_per_second": 32.367,
|
| 1802 |
+
"eval_steps_per_second": 2.027,
|
| 1803 |
+
"step": 78000
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 1.33845,
|
| 1807 |
+
"grad_norm": 0.7075666189193726,
|
| 1808 |
+
"learning_rate": 1.0763e-05,
|
| 1809 |
+
"loss": 5.6793,
|
| 1810 |
+
"step": 78500
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 1.34345,
|
| 1814 |
+
"grad_norm": 0.7042005658149719,
|
| 1815 |
+
"learning_rate": 1.0513e-05,
|
| 1816 |
+
"loss": 5.6636,
|
| 1817 |
+
"step": 79000
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 1.34345,
|
| 1821 |
+
"eval_accuracy": 0.03255183670892414,
|
| 1822 |
+
"eval_loss": 5.592087268829346,
|
| 1823 |
+
"eval_runtime": 111.6356,
|
| 1824 |
+
"eval_samples_per_second": 32.176,
|
| 1825 |
+
"eval_steps_per_second": 2.015,
|
| 1826 |
+
"step": 79000
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 1.34845,
|
| 1830 |
+
"grad_norm": 0.7470083832740784,
|
| 1831 |
+
"learning_rate": 1.0263e-05,
|
| 1832 |
+
"loss": 5.6727,
|
| 1833 |
+
"step": 79500
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 1.35345,
|
| 1837 |
+
"grad_norm": 0.7401617169380188,
|
| 1838 |
+
"learning_rate": 1.00135e-05,
|
| 1839 |
+
"loss": 5.6723,
|
| 1840 |
+
"step": 80000
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 1.35345,
|
| 1844 |
+
"eval_accuracy": 0.03255442201307565,
|
| 1845 |
+
"eval_loss": 5.588088512420654,
|
| 1846 |
+
"eval_runtime": 112.4329,
|
| 1847 |
+
"eval_samples_per_second": 31.948,
|
| 1848 |
+
"eval_steps_per_second": 2.001,
|
| 1849 |
+
"step": 80000
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 1.35845,
|
| 1853 |
+
"grad_norm": 0.7123810648918152,
|
| 1854 |
+
"learning_rate": 9.7635e-06,
|
| 1855 |
+
"loss": 5.6695,
|
| 1856 |
+
"step": 80500
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 1.36345,
|
| 1860 |
+
"grad_norm": 0.7234753966331482,
|
| 1861 |
+
"learning_rate": 9.5135e-06,
|
| 1862 |
+
"loss": 5.659,
|
| 1863 |
+
"step": 81000
|
| 1864 |
+
},
|
| 1865 |
+
{
|
| 1866 |
+
"epoch": 1.36345,
|
| 1867 |
+
"eval_accuracy": 0.03243753905169955,
|
| 1868 |
+
"eval_loss": 5.5822529792785645,
|
| 1869 |
+
"eval_runtime": 115.5372,
|
| 1870 |
+
"eval_samples_per_second": 31.09,
|
| 1871 |
+
"eval_steps_per_second": 1.947,
|
| 1872 |
+
"step": 81000
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 1.36845,
|
| 1876 |
+
"grad_norm": 0.7145719528198242,
|
| 1877 |
+
"learning_rate": 9.2635e-06,
|
| 1878 |
+
"loss": 5.6632,
|
| 1879 |
+
"step": 81500
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 1.37345,
|
| 1883 |
+
"grad_norm": 0.7814493179321289,
|
| 1884 |
+
"learning_rate": 9.013500000000001e-06,
|
| 1885 |
+
"loss": 5.6729,
|
| 1886 |
+
"step": 82000
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 1.37345,
|
| 1890 |
+
"eval_accuracy": 0.032616741449990966,
|
| 1891 |
+
"eval_loss": 5.579476833343506,
|
| 1892 |
+
"eval_runtime": 111.3925,
|
| 1893 |
+
"eval_samples_per_second": 32.246,
|
| 1894 |
+
"eval_steps_per_second": 2.02,
|
| 1895 |
+
"step": 82000
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 1.37845,
|
| 1899 |
+
"grad_norm": 0.7196946144104004,
|
| 1900 |
+
"learning_rate": 8.7635e-06,
|
| 1901 |
+
"loss": 5.6638,
|
| 1902 |
+
"step": 82500
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"epoch": 1.38345,
|
| 1906 |
+
"grad_norm": 0.7334076762199402,
|
| 1907 |
+
"learning_rate": 8.514e-06,
|
| 1908 |
+
"loss": 5.6595,
|
| 1909 |
+
"step": 83000
|
| 1910 |
+
},
|
| 1911 |
+
{
|
| 1912 |
+
"epoch": 1.38345,
|
| 1913 |
+
"eval_accuracy": 0.03224581833856925,
|
| 1914 |
+
"eval_loss": 5.579442977905273,
|
| 1915 |
+
"eval_runtime": 111.9776,
|
| 1916 |
+
"eval_samples_per_second": 32.078,
|
| 1917 |
+
"eval_steps_per_second": 2.009,
|
| 1918 |
+
"step": 83000
|
| 1919 |
+
},
|
| 1920 |
+
{
|
| 1921 |
+
"epoch": 1.38845,
|
| 1922 |
+
"grad_norm": 0.797461748123169,
|
| 1923 |
+
"learning_rate": 8.264e-06,
|
| 1924 |
+
"loss": 5.66,
|
| 1925 |
+
"step": 83500
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 1.39345,
|
| 1929 |
+
"grad_norm": 0.7179501056671143,
|
| 1930 |
+
"learning_rate": 8.014e-06,
|
| 1931 |
+
"loss": 5.6565,
|
| 1932 |
+
"step": 84000
|
| 1933 |
+
},
|
| 1934 |
+
{
|
| 1935 |
+
"epoch": 1.39345,
|
| 1936 |
+
"eval_accuracy": 0.032768049777174,
|
| 1937 |
+
"eval_loss": 5.575778961181641,
|
| 1938 |
+
"eval_runtime": 113.2284,
|
| 1939 |
+
"eval_samples_per_second": 31.723,
|
| 1940 |
+
"eval_steps_per_second": 1.987,
|
| 1941 |
+
"step": 84000
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 1.39845,
|
| 1945 |
+
"grad_norm": 0.7444576025009155,
|
| 1946 |
+
"learning_rate": 7.764e-06,
|
| 1947 |
+
"loss": 5.6539,
|
| 1948 |
+
"step": 84500
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 1.4034499999999999,
|
| 1952 |
+
"grad_norm": 0.681348979473114,
|
| 1953 |
+
"learning_rate": 7.514500000000001e-06,
|
| 1954 |
+
"loss": 5.6649,
|
| 1955 |
+
"step": 85000
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 1.4034499999999999,
|
| 1959 |
+
"eval_accuracy": 0.03250598157739475,
|
| 1960 |
+
"eval_loss": 5.571649074554443,
|
| 1961 |
+
"eval_runtime": 110.7941,
|
| 1962 |
+
"eval_samples_per_second": 32.42,
|
| 1963 |
+
"eval_steps_per_second": 2.031,
|
| 1964 |
+
"step": 85000
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 1.40845,
|
| 1968 |
+
"grad_norm": 0.6666921973228455,
|
| 1969 |
+
"learning_rate": 7.2645000000000005e-06,
|
| 1970 |
+
"loss": 5.6487,
|
| 1971 |
+
"step": 85500
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 1.41345,
|
| 1975 |
+
"grad_norm": 0.7330692410469055,
|
| 1976 |
+
"learning_rate": 7.0145e-06,
|
| 1977 |
+
"loss": 5.6561,
|
| 1978 |
+
"step": 86000
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 1.41345,
|
| 1982 |
+
"eval_accuracy": 0.032128118965355834,
|
| 1983 |
+
"eval_loss": 5.5695481300354,
|
| 1984 |
+
"eval_runtime": 113.9227,
|
| 1985 |
+
"eval_samples_per_second": 31.53,
|
| 1986 |
+
"eval_steps_per_second": 1.975,
|
| 1987 |
+
"step": 86000
|
| 1988 |
+
},
|
| 1989 |
+
{
|
| 1990 |
+
"epoch": 1.41845,
|
| 1991 |
+
"grad_norm": 0.7140607237815857,
|
| 1992 |
+
"learning_rate": 6.7645e-06,
|
| 1993 |
+
"loss": 5.6569,
|
| 1994 |
+
"step": 86500
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 1.4234499999999999,
|
| 1998 |
+
"grad_norm": 0.695405125617981,
|
| 1999 |
+
"learning_rate": 6.5145e-06,
|
| 2000 |
+
"loss": 5.6405,
|
| 2001 |
+
"step": 87000
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"epoch": 1.4234499999999999,
|
| 2005 |
+
"eval_accuracy": 0.03226895000729328,
|
| 2006 |
+
"eval_loss": 5.565379619598389,
|
| 2007 |
+
"eval_runtime": 110.9885,
|
| 2008 |
+
"eval_samples_per_second": 32.364,
|
| 2009 |
+
"eval_steps_per_second": 2.027,
|
| 2010 |
+
"step": 87000
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"epoch": 1.42845,
|
| 2014 |
+
"grad_norm": 0.748406708240509,
|
| 2015 |
+
"learning_rate": 6.265e-06,
|
| 2016 |
+
"loss": 5.661,
|
| 2017 |
+
"step": 87500
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 1.4334500000000001,
|
| 2021 |
+
"grad_norm": 0.7667502760887146,
|
| 2022 |
+
"learning_rate": 6.015000000000001e-06,
|
| 2023 |
+
"loss": 5.6482,
|
| 2024 |
+
"step": 88000
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 1.4334500000000001,
|
| 2028 |
+
"eval_accuracy": 0.03212390083752969,
|
| 2029 |
+
"eval_loss": 5.562798023223877,
|
| 2030 |
+
"eval_runtime": 113.7863,
|
| 2031 |
+
"eval_samples_per_second": 31.568,
|
| 2032 |
+
"eval_steps_per_second": 1.977,
|
| 2033 |
+
"step": 88000
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 1.43845,
|
| 2037 |
+
"grad_norm": 0.7025783658027649,
|
| 2038 |
+
"learning_rate": 5.765e-06,
|
| 2039 |
+
"loss": 5.6537,
|
| 2040 |
+
"step": 88500
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 1.44345,
|
| 2044 |
+
"grad_norm": 0.7218544483184814,
|
| 2045 |
+
"learning_rate": 5.515e-06,
|
| 2046 |
+
"loss": 5.6425,
|
| 2047 |
+
"step": 89000
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 1.44345,
|
| 2051 |
+
"eval_accuracy": 0.03228555038131876,
|
| 2052 |
+
"eval_loss": 5.562201023101807,
|
| 2053 |
+
"eval_runtime": 110.7556,
|
| 2054 |
+
"eval_samples_per_second": 32.432,
|
| 2055 |
+
"eval_steps_per_second": 2.032,
|
| 2056 |
+
"step": 89000
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"epoch": 2.0019,
|
| 2060 |
+
"grad_norm": 0.7732229232788086,
|
| 2061 |
+
"learning_rate": 5.265e-06,
|
| 2062 |
+
"loss": 5.6439,
|
| 2063 |
+
"step": 89500
|
| 2064 |
+
},
|
| 2065 |
+
{
|
| 2066 |
+
"epoch": 2.0069,
|
| 2067 |
+
"grad_norm": 0.7334641814231873,
|
| 2068 |
+
"learning_rate": 5.015e-06,
|
| 2069 |
+
"loss": 5.6379,
|
| 2070 |
+
"step": 90000
|
| 2071 |
+
},
|
| 2072 |
+
{
|
| 2073 |
+
"epoch": 2.0069,
|
| 2074 |
+
"eval_accuracy": 0.032302831098542,
|
| 2075 |
+
"eval_loss": 5.558170318603516,
|
| 2076 |
+
"eval_runtime": 110.8037,
|
| 2077 |
+
"eval_samples_per_second": 32.418,
|
| 2078 |
+
"eval_steps_per_second": 2.031,
|
| 2079 |
+
"step": 90000
|
| 2080 |
+
},
|
| 2081 |
+
{
|
| 2082 |
+
"epoch": 2.0119,
|
| 2083 |
+
"grad_norm": 0.7253163456916809,
|
| 2084 |
+
"learning_rate": 4.765e-06,
|
| 2085 |
+
"loss": 5.6416,
|
| 2086 |
+
"step": 90500
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"epoch": 2.0169,
|
| 2090 |
+
"grad_norm": 0.6948328614234924,
|
| 2091 |
+
"learning_rate": 4.515000000000001e-06,
|
| 2092 |
+
"loss": 5.6357,
|
| 2093 |
+
"step": 91000
|
| 2094 |
+
},
|
| 2095 |
+
{
|
| 2096 |
+
"epoch": 2.0169,
|
| 2097 |
+
"eval_accuracy": 0.03219384011825997,
|
| 2098 |
+
"eval_loss": 5.557282447814941,
|
| 2099 |
+
"eval_runtime": 112.1174,
|
| 2100 |
+
"eval_samples_per_second": 32.038,
|
| 2101 |
+
"eval_steps_per_second": 2.007,
|
| 2102 |
+
"step": 91000
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 2.0219,
|
| 2106 |
+
"grad_norm": 0.8627682328224182,
|
| 2107 |
+
"learning_rate": 4.2655e-06,
|
| 2108 |
+
"loss": 5.6417,
|
| 2109 |
+
"step": 91500
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 2.0269,
|
| 2113 |
+
"grad_norm": 0.6853375434875488,
|
| 2114 |
+
"learning_rate": 4.015500000000001e-06,
|
| 2115 |
+
"loss": 5.6381,
|
| 2116 |
+
"step": 92000
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 2.0269,
|
| 2120 |
+
"eval_accuracy": 0.032045117095228455,
|
| 2121 |
+
"eval_loss": 5.556839942932129,
|
| 2122 |
+
"eval_runtime": 112.6404,
|
| 2123 |
+
"eval_samples_per_second": 31.889,
|
| 2124 |
+
"eval_steps_per_second": 1.998,
|
| 2125 |
+
"step": 92000
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 2.0319,
|
| 2129 |
+
"grad_norm": 0.7689598798751831,
|
| 2130 |
+
"learning_rate": 3.7655e-06,
|
| 2131 |
+
"loss": 5.6349,
|
| 2132 |
+
"step": 92500
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 2.0369,
|
| 2136 |
+
"grad_norm": 0.714249849319458,
|
| 2137 |
+
"learning_rate": 3.516e-06,
|
| 2138 |
+
"loss": 5.6427,
|
| 2139 |
+
"step": 93000
|
| 2140 |
+
},
|
| 2141 |
+
{
|
| 2142 |
+
"epoch": 2.0369,
|
| 2143 |
+
"eval_accuracy": 0.032379982017168595,
|
| 2144 |
+
"eval_loss": 5.5526018142700195,
|
| 2145 |
+
"eval_runtime": 114.2105,
|
| 2146 |
+
"eval_samples_per_second": 31.451,
|
| 2147 |
+
"eval_steps_per_second": 1.97,
|
| 2148 |
+
"step": 93000
|
| 2149 |
+
},
|
| 2150 |
+
{
|
| 2151 |
+
"epoch": 2.0419,
|
| 2152 |
+
"grad_norm": 0.7157047390937805,
|
| 2153 |
+
"learning_rate": 3.266e-06,
|
| 2154 |
+
"loss": 5.6238,
|
| 2155 |
+
"step": 93500
|
| 2156 |
+
},
|
| 2157 |
+
{
|
| 2158 |
+
"epoch": 2.0469,
|
| 2159 |
+
"grad_norm": 0.702506959438324,
|
| 2160 |
+
"learning_rate": 3.016e-06,
|
| 2161 |
+
"loss": 5.6364,
|
| 2162 |
+
"step": 94000
|
| 2163 |
+
},
|
| 2164 |
+
{
|
| 2165 |
+
"epoch": 2.0469,
|
| 2166 |
+
"eval_accuracy": 0.03230364751037931,
|
| 2167 |
+
"eval_loss": 5.55258321762085,
|
| 2168 |
+
"eval_runtime": 113.8093,
|
| 2169 |
+
"eval_samples_per_second": 31.562,
|
| 2170 |
+
"eval_steps_per_second": 1.977,
|
| 2171 |
+
"step": 94000
|
| 2172 |
+
},
|
| 2173 |
+
{
|
| 2174 |
+
"epoch": 2.0519,
|
| 2175 |
+
"grad_norm": 0.7739561200141907,
|
| 2176 |
+
"learning_rate": 2.7660000000000003e-06,
|
| 2177 |
+
"loss": 5.6289,
|
| 2178 |
+
"step": 94500
|
| 2179 |
+
},
|
| 2180 |
+
{
|
| 2181 |
+
"epoch": 2.0569,
|
| 2182 |
+
"grad_norm": 0.700249195098877,
|
| 2183 |
+
"learning_rate": 2.516e-06,
|
| 2184 |
+
"loss": 5.626,
|
| 2185 |
+
"step": 95000
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 2.0569,
|
| 2189 |
+
"eval_accuracy": 0.032115600650516954,
|
| 2190 |
+
"eval_loss": 5.550052642822266,
|
| 2191 |
+
"eval_runtime": 112.9135,
|
| 2192 |
+
"eval_samples_per_second": 31.812,
|
| 2193 |
+
"eval_steps_per_second": 1.993,
|
| 2194 |
+
"step": 95000
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 2.0619,
|
| 2198 |
+
"grad_norm": 0.694694995880127,
|
| 2199 |
+
"learning_rate": 2.266e-06,
|
| 2200 |
+
"loss": 5.6414,
|
| 2201 |
+
"step": 95500
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 2.0669,
|
| 2205 |
+
"grad_norm": 0.8426607251167297,
|
| 2206 |
+
"learning_rate": 2.0165e-06,
|
| 2207 |
+
"loss": 5.636,
|
| 2208 |
+
"step": 96000
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 2.0669,
|
| 2212 |
+
"eval_accuracy": 0.032368688320085694,
|
| 2213 |
+
"eval_loss": 5.549187183380127,
|
| 2214 |
+
"eval_runtime": 111.6948,
|
| 2215 |
+
"eval_samples_per_second": 32.159,
|
| 2216 |
+
"eval_steps_per_second": 2.014,
|
| 2217 |
+
"step": 96000
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"epoch": 2.0719,
|
| 2221 |
+
"grad_norm": 0.7448583841323853,
|
| 2222 |
+
"learning_rate": 1.7665000000000002e-06,
|
| 2223 |
+
"loss": 5.633,
|
| 2224 |
+
"step": 96500
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"epoch": 2.0769,
|
| 2228 |
+
"grad_norm": 0.7047140002250671,
|
| 2229 |
+
"learning_rate": 1.5165e-06,
|
| 2230 |
+
"loss": 5.632,
|
| 2231 |
+
"step": 97000
|
| 2232 |
+
},
|
| 2233 |
+
{
|
| 2234 |
+
"epoch": 2.0769,
|
| 2235 |
+
"eval_accuracy": 0.032333582611081,
|
| 2236 |
+
"eval_loss": 5.548858165740967,
|
| 2237 |
+
"eval_runtime": 111.3384,
|
| 2238 |
+
"eval_samples_per_second": 32.262,
|
| 2239 |
+
"eval_steps_per_second": 2.021,
|
| 2240 |
+
"step": 97000
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 2.0819,
|
| 2244 |
+
"grad_norm": 0.7185714244842529,
|
| 2245 |
+
"learning_rate": 1.2665e-06,
|
| 2246 |
+
"loss": 5.6218,
|
| 2247 |
+
"step": 97500
|
| 2248 |
+
},
|
| 2249 |
+
{
|
| 2250 |
+
"epoch": 2.0869,
|
| 2251 |
+
"grad_norm": 0.906790018081665,
|
| 2252 |
+
"learning_rate": 1.0165000000000001e-06,
|
| 2253 |
+
"loss": 5.6133,
|
| 2254 |
+
"step": 98000
|
| 2255 |
+
},
|
| 2256 |
+
{
|
| 2257 |
+
"epoch": 2.0869,
|
| 2258 |
+
"eval_accuracy": 0.03233031696373172,
|
| 2259 |
+
"eval_loss": 5.547926425933838,
|
| 2260 |
+
"eval_runtime": 112.1516,
|
| 2261 |
+
"eval_samples_per_second": 32.028,
|
| 2262 |
+
"eval_steps_per_second": 2.006,
|
| 2263 |
+
"step": 98000
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 2.0919,
|
| 2267 |
+
"grad_norm": 0.7219076752662659,
|
| 2268 |
+
"learning_rate": 7.67e-07,
|
| 2269 |
+
"loss": 5.6305,
|
| 2270 |
+
"step": 98500
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 2.0969,
|
| 2274 |
+
"grad_norm": 0.7102829217910767,
|
| 2275 |
+
"learning_rate": 5.17e-07,
|
| 2276 |
+
"loss": 5.6291,
|
| 2277 |
+
"step": 99000
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 2.0969,
|
| 2281 |
+
"eval_accuracy": 0.032251125015511826,
|
| 2282 |
+
"eval_loss": 5.54772424697876,
|
| 2283 |
+
"eval_runtime": 113.02,
|
| 2284 |
+
"eval_samples_per_second": 31.782,
|
| 2285 |
+
"eval_steps_per_second": 1.991,
|
| 2286 |
+
"step": 99000
|
| 2287 |
+
},
|
| 2288 |
+
{
|
| 2289 |
+
"epoch": 2.1019,
|
| 2290 |
+
"grad_norm": 0.7246349453926086,
|
| 2291 |
+
"learning_rate": 2.67e-07,
|
| 2292 |
+
"loss": 5.6351,
|
| 2293 |
+
"step": 99500
|
| 2294 |
+
},
|
| 2295 |
+
{
|
| 2296 |
+
"epoch": 2.1069,
|
| 2297 |
+
"grad_norm": 0.6908907294273376,
|
| 2298 |
+
"learning_rate": 1.7000000000000003e-08,
|
| 2299 |
+
"loss": 5.6271,
|
| 2300 |
+
"step": 100000
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 2.1069,
|
| 2304 |
+
"eval_accuracy": 0.0322300343763811,
|
| 2305 |
+
"eval_loss": 5.546974182128906,
|
| 2306 |
+
"eval_runtime": 114.2737,
|
| 2307 |
+
"eval_samples_per_second": 31.433,
|
| 2308 |
+
"eval_steps_per_second": 1.969,
|
| 2309 |
+
"step": 100000
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"epoch": 1.00001,
|
| 2313 |
+
"step": 100001,
|
| 2314 |
+
"total_flos": 9.182126159167488e+17,
|
| 2315 |
+
"train_loss": 5.625401986489168e-05,
|
| 2316 |
+
"train_runtime": 26.8473,
|
| 2317 |
+
"train_samples_per_second": 59596.412,
|
| 2318 |
+
"train_steps_per_second": 3724.776
|
| 2319 |
+
}
|
| 2320 |
+
],
|
| 2321 |
+
"logging_steps": 500,
|
| 2322 |
+
"max_steps": 100000,
|
| 2323 |
+
"num_input_tokens_seen": 0,
|
| 2324 |
+
"num_train_epochs": 9223372036854775807,
|
| 2325 |
+
"save_steps": 500,
|
| 2326 |
+
"stateful_callbacks": {
|
| 2327 |
+
"TrainerControl": {
|
| 2328 |
+
"args": {
|
| 2329 |
+
"should_epoch_stop": false,
|
| 2330 |
+
"should_evaluate": false,
|
| 2331 |
+
"should_log": false,
|
| 2332 |
+
"should_save": true,
|
| 2333 |
+
"should_training_stop": true
|
| 2334 |
+
},
|
| 2335 |
+
"attributes": {}
|
| 2336 |
+
}
|
| 2337 |
+
},
|
| 2338 |
+
"total_flos": 9.182126159167488e+17,
|
| 2339 |
+
"train_batch_size": 8,
|
| 2340 |
+
"trial_name": null,
|
| 2341 |
+
"trial_params": null
|
| 2342 |
+
}
|