Commit
•
2ab8b96
1
Parent(s):
ed174e2
Model save
Browse files- README.md +64 -0
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +1407 -0
README.md
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: google/gemma-2-2b-it
|
| 3 |
+
datasets:
|
| 4 |
+
- generator
|
| 5 |
+
library_name: peft
|
| 6 |
+
license: gemma
|
| 7 |
+
tags:
|
| 8 |
+
- trl
|
| 9 |
+
- sft
|
| 10 |
+
- generated_from_trainer
|
| 11 |
+
model-index:
|
| 12 |
+
- name: gemma-2-2b-it-lora-sql
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# gemma-2-2b-it-lora-sql
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [google/gemma-2-2b-it](https://huggingface.co/google/gemma-2-2b-it) on the generator dataset.
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0002
|
| 41 |
+
- train_batch_size: 4
|
| 42 |
+
- eval_batch_size: 8
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 2
|
| 46 |
+
- gradient_accumulation_steps: 2
|
| 47 |
+
- total_train_batch_size: 16
|
| 48 |
+
- total_eval_batch_size: 16
|
| 49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 50 |
+
- lr_scheduler_type: cosine
|
| 51 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 52 |
+
- num_epochs: 3
|
| 53 |
+
|
| 54 |
+
### Training results
|
| 55 |
+
|
| 56 |
+
|
| 57 |
+
|
| 58 |
+
### Framework versions
|
| 59 |
+
|
| 60 |
+
- PEFT 0.12.0
|
| 61 |
+
- Transformers 4.44.2
|
| 62 |
+
- Pytorch 2.3.0+cu121
|
| 63 |
+
- Datasets 3.0.0
|
| 64 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9977046671767407,
|
| 3 |
+
"total_flos": 3660773289099264.0,
|
| 4 |
+
"train_loss": 0.4337311900958168,
|
| 5 |
+
"train_runtime": 5331.8775,
|
| 6 |
+
"train_samples": 100000,
|
| 7 |
+
"train_samples_per_second": 5.882,
|
| 8 |
+
"train_steps_per_second": 0.367
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9977046671767407,
|
| 3 |
+
"total_flos": 3660773289099264.0,
|
| 4 |
+
"train_loss": 0.4337311900958168,
|
| 5 |
+
"train_runtime": 5331.8775,
|
| 6 |
+
"train_samples": 100000,
|
| 7 |
+
"train_samples_per_second": 5.882,
|
| 8 |
+
"train_steps_per_second": 0.367
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1407 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9977046671767407,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1959,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.015302218821729151,
|
| 13 |
+
"grad_norm": 2.9595158525528062,
|
| 14 |
+
"learning_rate": 1.0204081632653061e-05,
|
| 15 |
+
"loss": 1.7025,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.030604437643458302,
|
| 20 |
+
"grad_norm": 1.263751808272248,
|
| 21 |
+
"learning_rate": 2.0408163265306123e-05,
|
| 22 |
+
"loss": 1.4094,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.045906656465187455,
|
| 27 |
+
"grad_norm": 1.1314499693827325,
|
| 28 |
+
"learning_rate": 3.061224489795919e-05,
|
| 29 |
+
"loss": 1.0742,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.061208875286916604,
|
| 34 |
+
"grad_norm": 0.43429385320294256,
|
| 35 |
+
"learning_rate": 4.0816326530612245e-05,
|
| 36 |
+
"loss": 0.8975,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.07651109410864575,
|
| 41 |
+
"grad_norm": 0.37910893323587813,
|
| 42 |
+
"learning_rate": 5.102040816326531e-05,
|
| 43 |
+
"loss": 0.7694,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.09181331293037491,
|
| 48 |
+
"grad_norm": 0.32985782659377816,
|
| 49 |
+
"learning_rate": 6.122448979591838e-05,
|
| 50 |
+
"loss": 0.6667,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.10711553175210406,
|
| 55 |
+
"grad_norm": 0.27993689719153514,
|
| 56 |
+
"learning_rate": 7.142857142857143e-05,
|
| 57 |
+
"loss": 0.6143,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.12241775057383321,
|
| 62 |
+
"grad_norm": 0.2562350918748776,
|
| 63 |
+
"learning_rate": 8.163265306122449e-05,
|
| 64 |
+
"loss": 0.5736,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.13771996939556236,
|
| 69 |
+
"grad_norm": 0.2627549443187762,
|
| 70 |
+
"learning_rate": 9.183673469387756e-05,
|
| 71 |
+
"loss": 0.5605,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.1530221882172915,
|
| 76 |
+
"grad_norm": 0.25450163445897056,
|
| 77 |
+
"learning_rate": 0.00010204081632653062,
|
| 78 |
+
"loss": 0.5425,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.16832440703902066,
|
| 83 |
+
"grad_norm": 0.265185130228763,
|
| 84 |
+
"learning_rate": 0.00011224489795918367,
|
| 85 |
+
"loss": 0.5417,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.18362662586074982,
|
| 90 |
+
"grad_norm": 0.2600738826510011,
|
| 91 |
+
"learning_rate": 0.00012244897959183676,
|
| 92 |
+
"loss": 0.5349,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.19892884468247896,
|
| 97 |
+
"grad_norm": 0.27451283782332153,
|
| 98 |
+
"learning_rate": 0.0001326530612244898,
|
| 99 |
+
"loss": 0.5225,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.21423106350420812,
|
| 104 |
+
"grad_norm": 0.29276216948080613,
|
| 105 |
+
"learning_rate": 0.00014285714285714287,
|
| 106 |
+
"loss": 0.5148,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.22953328232593725,
|
| 111 |
+
"grad_norm": 0.25373117999248507,
|
| 112 |
+
"learning_rate": 0.0001530612244897959,
|
| 113 |
+
"loss": 0.5108,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.24483550114766642,
|
| 118 |
+
"grad_norm": 0.26479254050664824,
|
| 119 |
+
"learning_rate": 0.00016326530612244898,
|
| 120 |
+
"loss": 0.5061,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.26013771996939555,
|
| 125 |
+
"grad_norm": 0.27232811822105624,
|
| 126 |
+
"learning_rate": 0.00017346938775510205,
|
| 127 |
+
"loss": 0.5065,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.2754399387911247,
|
| 132 |
+
"grad_norm": 0.26418337252250673,
|
| 133 |
+
"learning_rate": 0.00018367346938775512,
|
| 134 |
+
"loss": 0.4967,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.2907421576128539,
|
| 139 |
+
"grad_norm": 0.2503450300406005,
|
| 140 |
+
"learning_rate": 0.00019387755102040816,
|
| 141 |
+
"loss": 0.4998,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.306044376434583,
|
| 146 |
+
"grad_norm": 0.2339540170507543,
|
| 147 |
+
"learning_rate": 0.0001999974597101728,
|
| 148 |
+
"loss": 0.4989,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.32134659525631215,
|
| 153 |
+
"grad_norm": 0.23446730330407825,
|
| 154 |
+
"learning_rate": 0.0001999688829317862,
|
| 155 |
+
"loss": 0.4895,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.3366488140780413,
|
| 160 |
+
"grad_norm": 0.23685036500213205,
|
| 161 |
+
"learning_rate": 0.00019990856311693857,
|
| 162 |
+
"loss": 0.4898,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.3519510328997705,
|
| 167 |
+
"grad_norm": 0.2396529148643123,
|
| 168 |
+
"learning_rate": 0.00019981651941893068,
|
| 169 |
+
"loss": 0.4881,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.36725325172149964,
|
| 174 |
+
"grad_norm": 0.22794906865901965,
|
| 175 |
+
"learning_rate": 0.0001996927810643216,
|
| 176 |
+
"loss": 0.4825,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.38255547054322875,
|
| 181 |
+
"grad_norm": 0.21749849325438408,
|
| 182 |
+
"learning_rate": 0.00019953738734364843,
|
| 183 |
+
"loss": 0.4905,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.3978576893649579,
|
| 188 |
+
"grad_norm": 0.21941306857106413,
|
| 189 |
+
"learning_rate": 0.00019935038759895038,
|
| 190 |
+
"loss": 0.4844,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.4131599081866871,
|
| 195 |
+
"grad_norm": 0.2106883472247568,
|
| 196 |
+
"learning_rate": 0.0001991318412081012,
|
| 197 |
+
"loss": 0.4832,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.42846212700841624,
|
| 202 |
+
"grad_norm": 0.20643151525949935,
|
| 203 |
+
"learning_rate": 0.00019888181756595513,
|
| 204 |
+
"loss": 0.4732,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.44376434583014535,
|
| 209 |
+
"grad_norm": 0.21375448358620805,
|
| 210 |
+
"learning_rate": 0.0001986003960623118,
|
| 211 |
+
"loss": 0.4777,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.4590665646518745,
|
| 216 |
+
"grad_norm": 0.2078386748706879,
|
| 217 |
+
"learning_rate": 0.0001982876660567078,
|
| 218 |
+
"loss": 0.4773,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.4743687834736037,
|
| 223 |
+
"grad_norm": 0.20709767688640485,
|
| 224 |
+
"learning_rate": 0.00019794372685004232,
|
| 225 |
+
"loss": 0.4774,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.48967100229533284,
|
| 230 |
+
"grad_norm": 0.20114204358521104,
|
| 231 |
+
"learning_rate": 0.00019756868765304637,
|
| 232 |
+
"loss": 0.4724,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.504973221117062,
|
| 237 |
+
"grad_norm": 0.2047995080216857,
|
| 238 |
+
"learning_rate": 0.000197162667551605,
|
| 239 |
+
"loss": 0.4723,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.5202754399387911,
|
| 244 |
+
"grad_norm": 0.2065404899680565,
|
| 245 |
+
"learning_rate": 0.00019672579546894418,
|
| 246 |
+
"loss": 0.4744,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.5355776587605203,
|
| 251 |
+
"grad_norm": 0.20339328597837086,
|
| 252 |
+
"learning_rate": 0.00019625821012469392,
|
| 253 |
+
"loss": 0.469,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.5508798775822494,
|
| 258 |
+
"grad_norm": 0.20269051264889382,
|
| 259 |
+
"learning_rate": 0.0001957600599908406,
|
| 260 |
+
"loss": 0.4652,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.5661820964039785,
|
| 265 |
+
"grad_norm": 0.20140204074578216,
|
| 266 |
+
"learning_rate": 0.00019523150324458297,
|
| 267 |
+
"loss": 0.4663,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.5814843152257078,
|
| 272 |
+
"grad_norm": 0.20145479660046747,
|
| 273 |
+
"learning_rate": 0.0001946727077181062,
|
| 274 |
+
"loss": 0.4647,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.5967865340474369,
|
| 279 |
+
"grad_norm": 0.19812410832421173,
|
| 280 |
+
"learning_rate": 0.00019408385084529014,
|
| 281 |
+
"loss": 0.4669,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.612088752869166,
|
| 286 |
+
"grad_norm": 0.20593747645685082,
|
| 287 |
+
"learning_rate": 0.0001934651196053692,
|
| 288 |
+
"loss": 0.4606,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.6273909716908952,
|
| 293 |
+
"grad_norm": 0.19985586154896637,
|
| 294 |
+
"learning_rate": 0.00019281671046356057,
|
| 295 |
+
"loss": 0.465,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.6426931905126243,
|
| 300 |
+
"grad_norm": 0.18804149710765375,
|
| 301 |
+
"learning_rate": 0.0001921388293086812,
|
| 302 |
+
"loss": 0.4591,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.6579954093343535,
|
| 307 |
+
"grad_norm": 0.20403982281128524,
|
| 308 |
+
"learning_rate": 0.00019143169138777176,
|
| 309 |
+
"loss": 0.4612,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.6732976281560826,
|
| 314 |
+
"grad_norm": 0.19472172680549107,
|
| 315 |
+
"learning_rate": 0.00019069552123774966,
|
| 316 |
+
"loss": 0.4535,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.6885998469778117,
|
| 321 |
+
"grad_norm": 0.18853093267297413,
|
| 322 |
+
"learning_rate": 0.00018993055261411188,
|
| 323 |
+
"loss": 0.4536,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.703902065799541,
|
| 328 |
+
"grad_norm": 0.1824795154817547,
|
| 329 |
+
"learning_rate": 0.0001891370284167108,
|
| 330 |
+
"loss": 0.4533,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.7192042846212701,
|
| 335 |
+
"grad_norm": 0.19623573820167492,
|
| 336 |
+
"learning_rate": 0.00018831520061262657,
|
| 337 |
+
"loss": 0.4608,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.7345065034429993,
|
| 342 |
+
"grad_norm": 0.1894732171051352,
|
| 343 |
+
"learning_rate": 0.00018746533015615997,
|
| 344 |
+
"loss": 0.4561,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.7498087222647284,
|
| 349 |
+
"grad_norm": 0.1956746346911804,
|
| 350 |
+
"learning_rate": 0.00018658768690597198,
|
| 351 |
+
"loss": 0.4562,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.7651109410864575,
|
| 356 |
+
"grad_norm": 0.19485019146231133,
|
| 357 |
+
"learning_rate": 0.00018568254953939573,
|
| 358 |
+
"loss": 0.4547,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.7804131599081867,
|
| 363 |
+
"grad_norm": 0.1900932930656561,
|
| 364 |
+
"learning_rate": 0.0001847502054639483,
|
| 365 |
+
"loss": 0.4563,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.7957153787299158,
|
| 370 |
+
"grad_norm": 0.18608837489031835,
|
| 371 |
+
"learning_rate": 0.00018379095072607052,
|
| 372 |
+
"loss": 0.4546,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.811017597551645,
|
| 377 |
+
"grad_norm": 0.19229921508866923,
|
| 378 |
+
"learning_rate": 0.0001828050899171234,
|
| 379 |
+
"loss": 0.4506,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.8263198163733741,
|
| 384 |
+
"grad_norm": 0.20164999528455674,
|
| 385 |
+
"learning_rate": 0.00018179293607667178,
|
| 386 |
+
"loss": 0.4528,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.8416220351951033,
|
| 391 |
+
"grad_norm": 0.18962588415494633,
|
| 392 |
+
"learning_rate": 0.00018075481059308488,
|
| 393 |
+
"loss": 0.4501,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.8569242540168325,
|
| 398 |
+
"grad_norm": 0.1878318467581915,
|
| 399 |
+
"learning_rate": 0.00017969104310148627,
|
| 400 |
+
"loss": 0.4489,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.8722264728385616,
|
| 405 |
+
"grad_norm": 0.19855697462159105,
|
| 406 |
+
"learning_rate": 0.00017860197137908504,
|
| 407 |
+
"loss": 0.4486,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.8875286916602907,
|
| 412 |
+
"grad_norm": 0.18806947832467016,
|
| 413 |
+
"learning_rate": 0.00017748794123792169,
|
| 414 |
+
"loss": 0.449,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.9028309104820199,
|
| 419 |
+
"grad_norm": 0.18547154568221744,
|
| 420 |
+
"learning_rate": 0.00017634930641506272,
|
| 421 |
+
"loss": 0.4468,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.918133129303749,
|
| 426 |
+
"grad_norm": 0.1988415349574148,
|
| 427 |
+
"learning_rate": 0.00017518642846027876,
|
| 428 |
+
"loss": 0.4447,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.9334353481254782,
|
| 433 |
+
"grad_norm": 0.1971157063292002,
|
| 434 |
+
"learning_rate": 0.00017399967662124204,
|
| 435 |
+
"loss": 0.4496,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.9487375669472073,
|
| 440 |
+
"grad_norm": 0.18821213785618154,
|
| 441 |
+
"learning_rate": 0.00017278942772627954,
|
| 442 |
+
"loss": 0.4442,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.9640397857689365,
|
| 447 |
+
"grad_norm": 0.1863099123502169,
|
| 448 |
+
"learning_rate": 0.00017155606606471873,
|
| 449 |
+
"loss": 0.4428,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.9793420045906657,
|
| 454 |
+
"grad_norm": 0.1877422977434543,
|
| 455 |
+
"learning_rate": 0.00017029998326486485,
|
| 456 |
+
"loss": 0.4431,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.9946442234123948,
|
| 461 |
+
"grad_norm": 0.1976160670120698,
|
| 462 |
+
"learning_rate": 0.00016902157816964724,
|
| 463 |
+
"loss": 0.4382,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.009946442234124,
|
| 468 |
+
"grad_norm": 0.18992306538475748,
|
| 469 |
+
"learning_rate": 0.0001677212567099752,
|
| 470 |
+
"loss": 0.4353,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.025248661055853,
|
| 475 |
+
"grad_norm": 0.19209926834930846,
|
| 476 |
+
"learning_rate": 0.00016639943177584302,
|
| 477 |
+
"loss": 0.422,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.0405508798775822,
|
| 482 |
+
"grad_norm": 0.1862611696289539,
|
| 483 |
+
"learning_rate": 0.00016505652308522546,
|
| 484 |
+
"loss": 0.4166,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.0558530986993113,
|
| 489 |
+
"grad_norm": 0.19413670929825885,
|
| 490 |
+
"learning_rate": 0.00016369295705080493,
|
| 491 |
+
"loss": 0.4214,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.0711553175210407,
|
| 496 |
+
"grad_norm": 0.19044466403960753,
|
| 497 |
+
"learning_rate": 0.00016230916664457303,
|
| 498 |
+
"loss": 0.4226,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.0864575363427698,
|
| 503 |
+
"grad_norm": 0.20537259284486634,
|
| 504 |
+
"learning_rate": 0.00016090559126034955,
|
| 505 |
+
"loss": 0.4206,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.1017597551644989,
|
| 510 |
+
"grad_norm": 0.19548453583302455,
|
| 511 |
+
"learning_rate": 0.00015948267657426172,
|
| 512 |
+
"loss": 0.4258,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.117061973986228,
|
| 517 |
+
"grad_norm": 0.19233863084054012,
|
| 518 |
+
"learning_rate": 0.00015804087440322937,
|
| 519 |
+
"loss": 0.4212,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.132364192807957,
|
| 524 |
+
"grad_norm": 0.20059506523438514,
|
| 525 |
+
"learning_rate": 0.00015658064256149972,
|
| 526 |
+
"loss": 0.4248,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.1476664116296864,
|
| 531 |
+
"grad_norm": 0.19781190556296033,
|
| 532 |
+
"learning_rate": 0.00015510244471527798,
|
| 533 |
+
"loss": 0.419,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.1629686304514155,
|
| 538 |
+
"grad_norm": 0.1978522622616877,
|
| 539 |
+
"learning_rate": 0.0001536067502355,
|
| 540 |
+
"loss": 0.4156,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.1782708492731446,
|
| 545 |
+
"grad_norm": 0.19356090367390197,
|
| 546 |
+
"learning_rate": 0.00015209403404879303,
|
| 547 |
+
"loss": 0.4176,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.1935730680948737,
|
| 552 |
+
"grad_norm": 0.21041098882140832,
|
| 553 |
+
"learning_rate": 0.0001505647764866729,
|
| 554 |
+
"loss": 0.4155,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.2088752869166028,
|
| 559 |
+
"grad_norm": 0.19309297495512096,
|
| 560 |
+
"learning_rate": 0.00014901946313302452,
|
| 561 |
+
"loss": 0.4125,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.2241775057383322,
|
| 566 |
+
"grad_norm": 0.20988205146033515,
|
| 567 |
+
"learning_rate": 0.0001474585846699151,
|
| 568 |
+
"loss": 0.4198,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.2394797245600613,
|
| 573 |
+
"grad_norm": 0.19899429014402967,
|
| 574 |
+
"learning_rate": 0.00014588263672178812,
|
| 575 |
+
"loss": 0.4139,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.2547819433817904,
|
| 580 |
+
"grad_norm": 0.19678065811835893,
|
| 581 |
+
"learning_rate": 0.00014429211969808808,
|
| 582 |
+
"loss": 0.4168,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.2700841622035195,
|
| 587 |
+
"grad_norm": 0.20776458272762527,
|
| 588 |
+
"learning_rate": 0.00014268753863436602,
|
| 589 |
+
"loss": 0.4137,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.2853863810252486,
|
| 594 |
+
"grad_norm": 0.2056945302468653,
|
| 595 |
+
"learning_rate": 0.00014106940303191583,
|
| 596 |
+
"loss": 0.4166,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.300688599846978,
|
| 601 |
+
"grad_norm": 0.20731916359275016,
|
| 602 |
+
"learning_rate": 0.000139438226695993,
|
| 603 |
+
"loss": 0.4179,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.315990818668707,
|
| 608 |
+
"grad_norm": 0.2001144508118885,
|
| 609 |
+
"learning_rate": 0.00013779452757266617,
|
| 610 |
+
"loss": 0.4131,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.3312930374904361,
|
| 615 |
+
"grad_norm": 0.19624161193905473,
|
| 616 |
+
"learning_rate": 0.00013613882758435435,
|
| 617 |
+
"loss": 0.4089,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.3465952563121653,
|
| 622 |
+
"grad_norm": 0.20129114730483852,
|
| 623 |
+
"learning_rate": 0.0001344716524641012,
|
| 624 |
+
"loss": 0.4149,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.3618974751338944,
|
| 629 |
+
"grad_norm": 0.207327826679874,
|
| 630 |
+
"learning_rate": 0.0001327935315886395,
|
| 631 |
+
"loss": 0.4097,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.3771996939556237,
|
| 636 |
+
"grad_norm": 0.2029298678382067,
|
| 637 |
+
"learning_rate": 0.00013110499781029874,
|
| 638 |
+
"loss": 0.4132,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.3925019127773526,
|
| 643 |
+
"grad_norm": 0.20764777212748992,
|
| 644 |
+
"learning_rate": 0.00012940658728780862,
|
| 645 |
+
"loss": 0.4142,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.407804131599082,
|
| 650 |
+
"grad_norm": 0.2017977764168179,
|
| 651 |
+
"learning_rate": 0.00012769883931605333,
|
| 652 |
+
"loss": 0.4212,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.423106350420811,
|
| 657 |
+
"grad_norm": 0.1932903343458798,
|
| 658 |
+
"learning_rate": 0.00012598229615482954,
|
| 659 |
+
"loss": 0.4127,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.4384085692425401,
|
| 664 |
+
"grad_norm": 0.20197440137591363,
|
| 665 |
+
"learning_rate": 0.0001242575028566632,
|
| 666 |
+
"loss": 0.4118,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.4537107880642695,
|
| 671 |
+
"grad_norm": 0.20030727904644482,
|
| 672 |
+
"learning_rate": 0.00012252500709373934,
|
| 673 |
+
"loss": 0.4133,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.4690130068859983,
|
| 678 |
+
"grad_norm": 0.20837354923941284,
|
| 679 |
+
"learning_rate": 0.00012078535898400019,
|
| 680 |
+
"loss": 0.4117,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.4843152257077277,
|
| 685 |
+
"grad_norm": 0.19925595673498087,
|
| 686 |
+
"learning_rate": 0.00011903911091646684,
|
| 687 |
+
"loss": 0.4083,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.4996174445294568,
|
| 692 |
+
"grad_norm": 0.2084484154721651,
|
| 693 |
+
"learning_rate": 0.00011728681737583945,
|
| 694 |
+
"loss": 0.408,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.5149196633511859,
|
| 699 |
+
"grad_norm": 0.20623569681522963,
|
| 700 |
+
"learning_rate": 0.00011552903476643222,
|
| 701 |
+
"loss": 0.4117,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.5302218821729152,
|
| 706 |
+
"grad_norm": 0.19912897780166905,
|
| 707 |
+
"learning_rate": 0.0001137663212354988,
|
| 708 |
+
"loss": 0.4071,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.545524100994644,
|
| 713 |
+
"grad_norm": 0.2015032395684377,
|
| 714 |
+
"learning_rate": 0.00011199923649600432,
|
| 715 |
+
"loss": 0.4085,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.5608263198163734,
|
| 720 |
+
"grad_norm": 0.19846674785205634,
|
| 721 |
+
"learning_rate": 0.0001102283416489001,
|
| 722 |
+
"loss": 0.4094,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.5761285386381025,
|
| 727 |
+
"grad_norm": 0.19503368149396527,
|
| 728 |
+
"learning_rate": 0.00010845419900495772,
|
| 729 |
+
"loss": 0.4031,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.5914307574598316,
|
| 734 |
+
"grad_norm": 0.20636304448971973,
|
| 735 |
+
"learning_rate": 0.00010667737190621911,
|
| 736 |
+
"loss": 0.4141,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.606732976281561,
|
| 741 |
+
"grad_norm": 0.19741819850815578,
|
| 742 |
+
"learning_rate": 0.0001048984245471188,
|
| 743 |
+
"loss": 0.4123,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.6220351951032899,
|
| 748 |
+
"grad_norm": 0.208094236192064,
|
| 749 |
+
"learning_rate": 0.00010311792179533589,
|
| 750 |
+
"loss": 0.4094,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.6373374139250192,
|
| 755 |
+
"grad_norm": 0.20834394157642375,
|
| 756 |
+
"learning_rate": 0.00010133642901243199,
|
| 757 |
+
"loss": 0.4073,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.6526396327467483,
|
| 762 |
+
"grad_norm": 0.20325976490051528,
|
| 763 |
+
"learning_rate": 9.955451187433249e-05,
|
| 764 |
+
"loss": 0.4072,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.6679418515684774,
|
| 769 |
+
"grad_norm": 0.19996940436744937,
|
| 770 |
+
"learning_rate": 9.777273619170796e-05,
|
| 771 |
+
"loss": 0.4067,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.6832440703902067,
|
| 776 |
+
"grad_norm": 0.19609861659837347,
|
| 777 |
+
"learning_rate": 9.599166773031269e-05,
|
| 778 |
+
"loss": 0.405,
|
| 779 |
+
"step": 1100
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.6985462892119356,
|
| 783 |
+
"grad_norm": 0.20661663114073753,
|
| 784 |
+
"learning_rate": 9.421187203133763e-05,
|
| 785 |
+
"loss": 0.4089,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.713848508033665,
|
| 790 |
+
"grad_norm": 0.19921721692896047,
|
| 791 |
+
"learning_rate": 9.243391423183448e-05,
|
| 792 |
+
"loss": 0.4071,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.729150726855394,
|
| 797 |
+
"grad_norm": 0.20297141939912075,
|
| 798 |
+
"learning_rate": 9.06583588852683e-05,
|
| 799 |
+
"loss": 0.4043,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.7444529456771232,
|
| 804 |
+
"grad_norm": 0.20714391729089626,
|
| 805 |
+
"learning_rate": 8.888576978225527e-05,
|
| 806 |
+
"loss": 0.4031,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.7597551644988525,
|
| 811 |
+
"grad_norm": 0.20284798828869055,
|
| 812 |
+
"learning_rate": 8.711670977154274e-05,
|
| 813 |
+
"loss": 0.4051,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.7750573833205814,
|
| 818 |
+
"grad_norm": 0.20621816413716504,
|
| 819 |
+
"learning_rate": 8.535174058128812e-05,
|
| 820 |
+
"loss": 0.4028,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.7903596021423107,
|
| 825 |
+
"grad_norm": 0.2107861816265634,
|
| 826 |
+
"learning_rate": 8.359142264069424e-05,
|
| 827 |
+
"loss": 0.3972,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.8056618209640398,
|
| 832 |
+
"grad_norm": 0.20926246575879495,
|
| 833 |
+
"learning_rate": 8.183631490205637e-05,
|
| 834 |
+
"loss": 0.4033,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.820964039785769,
|
| 839 |
+
"grad_norm": 0.20398570550310058,
|
| 840 |
+
"learning_rate": 8.008697466327865e-05,
|
| 841 |
+
"loss": 0.3994,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.836266258607498,
|
| 846 |
+
"grad_norm": 0.21408610304309247,
|
| 847 |
+
"learning_rate": 7.834395739091585e-05,
|
| 848 |
+
"loss": 0.3971,
|
| 849 |
+
"step": 1200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.8515684774292271,
|
| 853 |
+
"grad_norm": 0.2059678534725396,
|
| 854 |
+
"learning_rate": 7.660781654379638e-05,
|
| 855 |
+
"loss": 0.4004,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.8668706962509565,
|
| 860 |
+
"grad_norm": 0.20089108656511492,
|
| 861 |
+
"learning_rate": 7.487910339728308e-05,
|
| 862 |
+
"loss": 0.398,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.8821729150726856,
|
| 867 |
+
"grad_norm": 0.2089057556225053,
|
| 868 |
+
"learning_rate": 7.315836686822729e-05,
|
| 869 |
+
"loss": 0.4,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.8974751338944147,
|
| 874 |
+
"grad_norm": 0.2076979091587327,
|
| 875 |
+
"learning_rate": 7.14461533406714e-05,
|
| 876 |
+
"loss": 0.3942,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.9127773527161438,
|
| 881 |
+
"grad_norm": 0.2075858197523019,
|
| 882 |
+
"learning_rate": 6.974300649235633e-05,
|
| 883 |
+
"loss": 0.4004,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.928079571537873,
|
| 888 |
+
"grad_norm": 0.21210310960815237,
|
| 889 |
+
"learning_rate": 6.804946712208793e-05,
|
| 890 |
+
"loss": 0.4021,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.9433817903596022,
|
| 895 |
+
"grad_norm": 0.21341887577251997,
|
| 896 |
+
"learning_rate": 6.63660729780174e-05,
|
| 897 |
+
"loss": 0.3954,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.9586840091813313,
|
| 902 |
+
"grad_norm": 0.22172559452313256,
|
| 903 |
+
"learning_rate": 6.469335858689074e-05,
|
| 904 |
+
"loss": 0.4002,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.9739862280030605,
|
| 909 |
+
"grad_norm": 0.20709389048082577,
|
| 910 |
+
"learning_rate": 6.303185508432085e-05,
|
| 911 |
+
"loss": 0.4018,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.9892884468247896,
|
| 916 |
+
"grad_norm": 0.21370736036441443,
|
| 917 |
+
"learning_rate": 6.138209004613647e-05,
|
| 918 |
+
"loss": 0.3955,
|
| 919 |
+
"step": 1300
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.0045906656465187,
|
| 923 |
+
"grad_norm": 0.20094204383719153,
|
| 924 |
+
"learning_rate": 5.974458732086149e-05,
|
| 925 |
+
"loss": 0.3851,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.019892884468248,
|
| 930 |
+
"grad_norm": 0.22289055701062488,
|
| 931 |
+
"learning_rate": 5.81198668633778e-05,
|
| 932 |
+
"loss": 0.3674,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.035195103289977,
|
| 937 |
+
"grad_norm": 0.22724332895866212,
|
| 938 |
+
"learning_rate": 5.6508444569824315e-05,
|
| 939 |
+
"loss": 0.3614,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.050497322111706,
|
| 944 |
+
"grad_norm": 0.23888425948419684,
|
| 945 |
+
"learning_rate": 5.491083211378505e-05,
|
| 946 |
+
"loss": 0.3614,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.0657995409334355,
|
| 951 |
+
"grad_norm": 0.23068081709857463,
|
| 952 |
+
"learning_rate": 5.3327536783817766e-05,
|
| 953 |
+
"loss": 0.3644,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.0811017597551644,
|
| 958 |
+
"grad_norm": 0.2460582871625401,
|
| 959 |
+
"learning_rate": 5.1759061322375045e-05,
|
| 960 |
+
"loss": 0.3634,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.0964039785768938,
|
| 965 |
+
"grad_norm": 0.24787239171921766,
|
| 966 |
+
"learning_rate": 5.0205903766168915e-05,
|
| 967 |
+
"loss": 0.3612,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.1117061973986226,
|
| 972 |
+
"grad_norm": 0.23563510103472043,
|
| 973 |
+
"learning_rate": 4.8668557288029684e-05,
|
| 974 |
+
"loss": 0.3627,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.127008416220352,
|
| 979 |
+
"grad_norm": 0.23966764070729765,
|
| 980 |
+
"learning_rate": 4.7147510040309115e-05,
|
| 981 |
+
"loss": 0.3633,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.1423106350420813,
|
| 986 |
+
"grad_norm": 0.24153476110732292,
|
| 987 |
+
"learning_rate": 4.56432449998779e-05,
|
| 988 |
+
"loss": 0.3621,
|
| 989 |
+
"step": 1400
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.15761285386381,
|
| 993 |
+
"grad_norm": 0.23966649969853318,
|
| 994 |
+
"learning_rate": 4.4156239814766355e-05,
|
| 995 |
+
"loss": 0.3597,
|
| 996 |
+
"step": 1410
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.1729150726855395,
|
| 1000 |
+
"grad_norm": 0.2492626290666687,
|
| 1001 |
+
"learning_rate": 4.268696665249724e-05,
|
| 1002 |
+
"loss": 0.3663,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.1882172915072684,
|
| 1007 |
+
"grad_norm": 0.24762132725809086,
|
| 1008 |
+
"learning_rate": 4.1235892050158866e-05,
|
| 1009 |
+
"loss": 0.3604,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.2035195103289977,
|
| 1014 |
+
"grad_norm": 0.24689838661058983,
|
| 1015 |
+
"learning_rate": 3.9803476766265835e-05,
|
| 1016 |
+
"loss": 0.3648,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.218821729150727,
|
| 1021 |
+
"grad_norm": 0.24705256051578678,
|
| 1022 |
+
"learning_rate": 3.839017563445489e-05,
|
| 1023 |
+
"loss": 0.3625,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.234123947972456,
|
| 1028 |
+
"grad_norm": 0.2496443256139868,
|
| 1029 |
+
"learning_rate": 3.699643741906193e-05,
|
| 1030 |
+
"loss": 0.3545,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.2494261667941853,
|
| 1035 |
+
"grad_norm": 0.249005841241916,
|
| 1036 |
+
"learning_rate": 3.562270467262619e-05,
|
| 1037 |
+
"loss": 0.3601,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 2.264728385615914,
|
| 1042 |
+
"grad_norm": 0.24757649232051404,
|
| 1043 |
+
"learning_rate": 3.426941359536699e-05,
|
| 1044 |
+
"loss": 0.3576,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 2.2800306044376435,
|
| 1049 |
+
"grad_norm": 0.25451870114574876,
|
| 1050 |
+
"learning_rate": 3.293699389667734e-05,
|
| 1051 |
+
"loss": 0.3648,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.295332823259373,
|
| 1056 |
+
"grad_norm": 0.2453131567916715,
|
| 1057 |
+
"learning_rate": 3.1625868658678784e-05,
|
| 1058 |
+
"loss": 0.3625,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.3106350420811017,
|
| 1063 |
+
"grad_norm": 0.26123253732278273,
|
| 1064 |
+
"learning_rate": 3.0336454201880404e-05,
|
| 1065 |
+
"loss": 0.3557,
|
| 1066 |
+
"step": 1510
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 2.325937260902831,
|
| 1070 |
+
"grad_norm": 0.2626580334343511,
|
| 1071 |
+
"learning_rate": 2.9069159952984938e-05,
|
| 1072 |
+
"loss": 0.3566,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 2.34123947972456,
|
| 1077 |
+
"grad_norm": 0.25003501828393526,
|
| 1078 |
+
"learning_rate": 2.7824388314883876e-05,
|
| 1079 |
+
"loss": 0.3591,
|
| 1080 |
+
"step": 1530
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 2.3565416985462893,
|
| 1084 |
+
"grad_norm": 0.2507207739640118,
|
| 1085 |
+
"learning_rate": 2.6602534538882752e-05,
|
| 1086 |
+
"loss": 0.358,
|
| 1087 |
+
"step": 1540
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 2.371843917368018,
|
| 1091 |
+
"grad_norm": 0.2512257287691549,
|
| 1092 |
+
"learning_rate": 2.5403986599197403e-05,
|
| 1093 |
+
"loss": 0.3501,
|
| 1094 |
+
"step": 1550
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.3871461361897475,
|
| 1098 |
+
"grad_norm": 0.2499142391014787,
|
| 1099 |
+
"learning_rate": 2.4229125069760773e-05,
|
| 1100 |
+
"loss": 0.3517,
|
| 1101 |
+
"step": 1560
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.402448355011477,
|
| 1105 |
+
"grad_norm": 0.24884412260444377,
|
| 1106 |
+
"learning_rate": 2.30783230033796e-05,
|
| 1107 |
+
"loss": 0.3568,
|
| 1108 |
+
"step": 1570
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 2.4177505738332057,
|
| 1112 |
+
"grad_norm": 0.25335712662759613,
|
| 1113 |
+
"learning_rate": 2.1951945813279306e-05,
|
| 1114 |
+
"loss": 0.3547,
|
| 1115 |
+
"step": 1580
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.433052792654935,
|
| 1119 |
+
"grad_norm": 0.24937746984746909,
|
| 1120 |
+
"learning_rate": 2.0850351157074598e-05,
|
| 1121 |
+
"loss": 0.353,
|
| 1122 |
+
"step": 1590
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.4483550114766643,
|
| 1126 |
+
"grad_norm": 0.26237405083230225,
|
| 1127 |
+
"learning_rate": 1.9773888823202747e-05,
|
| 1128 |
+
"loss": 0.3579,
|
| 1129 |
+
"step": 1600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.4636572302983932,
|
| 1133 |
+
"grad_norm": 0.24937407019841143,
|
| 1134 |
+
"learning_rate": 1.8722900619855577e-05,
|
| 1135 |
+
"loss": 0.3562,
|
| 1136 |
+
"step": 1610
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.4789594491201226,
|
| 1140 |
+
"grad_norm": 0.2497739493217595,
|
| 1141 |
+
"learning_rate": 1.7697720266445374e-05,
|
| 1142 |
+
"loss": 0.3512,
|
| 1143 |
+
"step": 1620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.4942616679418514,
|
| 1147 |
+
"grad_norm": 0.250399923768968,
|
| 1148 |
+
"learning_rate": 1.6698673287639242e-05,
|
| 1149 |
+
"loss": 0.3556,
|
| 1150 |
+
"step": 1630
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.5095638867635808,
|
| 1154 |
+
"grad_norm": 0.2538234603963528,
|
| 1155 |
+
"learning_rate": 1.5726076909995525e-05,
|
| 1156 |
+
"loss": 0.355,
|
| 1157 |
+
"step": 1640
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.5248661055853097,
|
| 1161 |
+
"grad_norm": 0.2504434686740776,
|
| 1162 |
+
"learning_rate": 1.4780239961235143e-05,
|
| 1163 |
+
"loss": 0.3581,
|
| 1164 |
+
"step": 1650
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.540168324407039,
|
| 1168 |
+
"grad_norm": 0.2537006354023553,
|
| 1169 |
+
"learning_rate": 1.3861462772179735e-05,
|
| 1170 |
+
"loss": 0.3529,
|
| 1171 |
+
"step": 1660
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 2.5554705432287683,
|
| 1175 |
+
"grad_norm": 0.2560925692389684,
|
| 1176 |
+
"learning_rate": 1.297003708138792e-05,
|
| 1177 |
+
"loss": 0.3584,
|
| 1178 |
+
"step": 1670
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 2.570772762050497,
|
| 1182 |
+
"grad_norm": 0.25340005468102866,
|
| 1183 |
+
"learning_rate": 1.2106245942519745e-05,
|
| 1184 |
+
"loss": 0.3562,
|
| 1185 |
+
"step": 1680
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 2.5860749808722265,
|
| 1189 |
+
"grad_norm": 0.25785387176401525,
|
| 1190 |
+
"learning_rate": 1.1270363634458903e-05,
|
| 1191 |
+
"loss": 0.3541,
|
| 1192 |
+
"step": 1690
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 2.601377199693956,
|
| 1196 |
+
"grad_norm": 0.2615280811952495,
|
| 1197 |
+
"learning_rate": 1.0462655574221213e-05,
|
| 1198 |
+
"loss": 0.3553,
|
| 1199 |
+
"step": 1700
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 2.6166794185156848,
|
| 1203 |
+
"grad_norm": 0.24663819526398556,
|
| 1204 |
+
"learning_rate": 9.683378232676965e-06,
|
| 1205 |
+
"loss": 0.3532,
|
| 1206 |
+
"step": 1710
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 2.631981637337414,
|
| 1210 |
+
"grad_norm": 0.2614751928374674,
|
| 1211 |
+
"learning_rate": 8.932779053113893e-06,
|
| 1212 |
+
"loss": 0.3546,
|
| 1213 |
+
"step": 1720
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 2.647283856159143,
|
| 1217 |
+
"grad_norm": 0.25884670404819016,
|
| 1218 |
+
"learning_rate": 8.211096372666783e-06,
|
| 1219 |
+
"loss": 0.3559,
|
| 1220 |
+
"step": 1730
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 2.6625860749808723,
|
| 1224 |
+
"grad_norm": 0.26303763365287397,
|
| 1225 |
+
"learning_rate": 7.518559346638432e-06,
|
| 1226 |
+
"loss": 0.3556,
|
| 1227 |
+
"step": 1740
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 2.677888293802601,
|
| 1231 |
+
"grad_norm": 0.2591833975916801,
|
| 1232 |
+
"learning_rate": 6.855387875736152e-06,
|
| 1233 |
+
"loss": 0.3577,
|
| 1234 |
+
"step": 1750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 2.6931905126243305,
|
| 1238 |
+
"grad_norm": 0.24688502674320847,
|
| 1239 |
+
"learning_rate": 6.221792536246973e-06,
|
| 1240 |
+
"loss": 0.3557,
|
| 1241 |
+
"step": 1760
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 2.70849273144606,
|
| 1245 |
+
"grad_norm": 0.26100100849151864,
|
| 1246 |
+
"learning_rate": 5.617974513173341e-06,
|
| 1247 |
+
"loss": 0.3548,
|
| 1248 |
+
"step": 1770
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 2.7237949502677887,
|
| 1252 |
+
"grad_norm": 0.2699346579349574,
|
| 1253 |
+
"learning_rate": 5.044125536351196e-06,
|
| 1254 |
+
"loss": 0.3511,
|
| 1255 |
+
"step": 1780
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 2.739097169089518,
|
| 1259 |
+
"grad_norm": 0.2717440390716044,
|
| 1260 |
+
"learning_rate": 4.500427819570097e-06,
|
| 1261 |
+
"loss": 0.3494,
|
| 1262 |
+
"step": 1790
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 2.7543993879112474,
|
| 1266 |
+
"grad_norm": 0.2579283254483208,
|
| 1267 |
+
"learning_rate": 3.987054002714952e-06,
|
| 1268 |
+
"loss": 0.3569,
|
| 1269 |
+
"step": 1800
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 2.7697016067329763,
|
| 1273 |
+
"grad_norm": 0.2594723269046269,
|
| 1274 |
+
"learning_rate": 3.504167096947952e-06,
|
| 1275 |
+
"loss": 0.3516,
|
| 1276 |
+
"step": 1810
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 2.785003825554705,
|
| 1280 |
+
"grad_norm": 0.256455133609715,
|
| 1281 |
+
"learning_rate": 3.051920432947664e-06,
|
| 1282 |
+
"loss": 0.3512,
|
| 1283 |
+
"step": 1820
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 2.8003060443764345,
|
| 1287 |
+
"grad_norm": 0.2621246657840175,
|
| 1288 |
+
"learning_rate": 2.6304576122221035e-06,
|
| 1289 |
+
"loss": 0.3527,
|
| 1290 |
+
"step": 1830
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 2.815608263198164,
|
| 1294 |
+
"grad_norm": 0.26278995363196544,
|
| 1295 |
+
"learning_rate": 2.2399124615110846e-06,
|
| 1296 |
+
"loss": 0.3518,
|
| 1297 |
+
"step": 1840
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 2.8309104820198927,
|
| 1301 |
+
"grad_norm": 0.27091602282401345,
|
| 1302 |
+
"learning_rate": 1.880408990292315e-06,
|
| 1303 |
+
"loss": 0.3581,
|
| 1304 |
+
"step": 1850
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 2.846212700841622,
|
| 1308 |
+
"grad_norm": 0.25572786473847814,
|
| 1309 |
+
"learning_rate": 1.5520613514047655e-06,
|
| 1310 |
+
"loss": 0.3562,
|
| 1311 |
+
"step": 1860
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 2.8615149196633514,
|
| 1315 |
+
"grad_norm": 0.26244627787054337,
|
| 1316 |
+
"learning_rate": 1.2549738048017846e-06,
|
| 1317 |
+
"loss": 0.3552,
|
| 1318 |
+
"step": 1870
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 2.8768171384850802,
|
| 1322 |
+
"grad_norm": 0.26131103676147655,
|
| 1323 |
+
"learning_rate": 9.892406844456026e-07,
|
| 1324 |
+
"loss": 0.352,
|
| 1325 |
+
"step": 1880
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 2.8921193573068096,
|
| 1329 |
+
"grad_norm": 0.27449669022834305,
|
| 1330 |
+
"learning_rate": 7.549463683534374e-07,
|
| 1331 |
+
"loss": 0.3504,
|
| 1332 |
+
"step": 1890
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 2.907421576128539,
|
| 1336 |
+
"grad_norm": 0.264605800831863,
|
| 1337 |
+
"learning_rate": 5.521652518051368e-07,
|
| 1338 |
+
"loss": 0.3515,
|
| 1339 |
+
"step": 1900
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 2.922723794950268,
|
| 1343 |
+
"grad_norm": 0.2618398016329481,
|
| 1344 |
+
"learning_rate": 3.809617237203744e-07,
|
| 1345 |
+
"loss": 0.3561,
|
| 1346 |
+
"step": 1910
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 2.9380260137719967,
|
| 1350 |
+
"grad_norm": 0.26831015620903936,
|
| 1351 |
+
"learning_rate": 2.4139014621340494e-07,
|
| 1352 |
+
"loss": 0.3527,
|
| 1353 |
+
"step": 1920
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 2.953328232593726,
|
| 1357 |
+
"grad_norm": 0.26666302239559175,
|
| 1358 |
+
"learning_rate": 1.334948373314493e-07,
|
| 1359 |
+
"loss": 0.3562,
|
| 1360 |
+
"step": 1930
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 2.9686304514154553,
|
| 1364 |
+
"grad_norm": 0.262359494195635,
|
| 1365 |
+
"learning_rate": 5.7310056982418094e-08,
|
| 1366 |
+
"loss": 0.355,
|
| 1367 |
+
"step": 1940
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 2.9839326702371842,
|
| 1371 |
+
"grad_norm": 0.2576422839291312,
|
| 1372 |
+
"learning_rate": 1.2859996056402423e-08,
|
| 1373 |
+
"loss": 0.3531,
|
| 1374 |
+
"step": 1950
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 2.9977046671767407,
|
| 1378 |
+
"step": 1959,
|
| 1379 |
+
"total_flos": 3660773289099264.0,
|
| 1380 |
+
"train_loss": 0.4337311900958168,
|
| 1381 |
+
"train_runtime": 5331.8775,
|
| 1382 |
+
"train_samples_per_second": 5.882,
|
| 1383 |
+
"train_steps_per_second": 0.367
|
| 1384 |
+
}
|
| 1385 |
+
],
|
| 1386 |
+
"logging_steps": 10,
|
| 1387 |
+
"max_steps": 1959,
|
| 1388 |
+
"num_input_tokens_seen": 0,
|
| 1389 |
+
"num_train_epochs": 3,
|
| 1390 |
+
"save_steps": 500,
|
| 1391 |
+
"stateful_callbacks": {
|
| 1392 |
+
"TrainerControl": {
|
| 1393 |
+
"args": {
|
| 1394 |
+
"should_epoch_stop": false,
|
| 1395 |
+
"should_evaluate": false,
|
| 1396 |
+
"should_log": false,
|
| 1397 |
+
"should_save": true,
|
| 1398 |
+
"should_training_stop": true
|
| 1399 |
+
},
|
| 1400 |
+
"attributes": {}
|
| 1401 |
+
}
|
| 1402 |
+
},
|
| 1403 |
+
"total_flos": 3660773289099264.0,
|
| 1404 |
+
"train_batch_size": 4,
|
| 1405 |
+
"trial_name": null,
|
| 1406 |
+
"trial_params": null
|
| 1407 |
+
}
|