Commit
•
b61de6c
1
Parent(s):
0407a77
Model save
Browse files- README.md +67 -0
- all_results.json +9 -0
- generation_config.json +12 -0
- train_results.json +9 -0
- trainer_state.json +2088 -0
README.md
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: llama3.1
|
| 4 |
+
base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: google_cloude_test_20k_transduction-gpt4omini_lr1e-5_epoch2_seed100_3
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# google_cloude_test_20k_transduction-gpt4omini_lr1e-5_epoch2_seed100_3
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.0615
|
| 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: 1e-05
|
| 41 |
+
- train_batch_size: 8
|
| 42 |
+
- eval_batch_size: 4
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 8
|
| 46 |
+
- gradient_accumulation_steps: 2
|
| 47 |
+
- total_train_batch_size: 128
|
| 48 |
+
- total_eval_batch_size: 32
|
| 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: 2
|
| 53 |
+
|
| 54 |
+
### Training results
|
| 55 |
+
|
| 56 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 57 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 58 |
+
| 0.0963 | 0.9966 | 145 | 0.0735 |
|
| 59 |
+
| 0.0658 | 1.9931 | 290 | 0.0615 |
|
| 60 |
+
|
| 61 |
+
|
| 62 |
+
### Framework versions
|
| 63 |
+
|
| 64 |
+
- Transformers 4.45.0.dev0
|
| 65 |
+
- Pytorch 2.4.0+cu121
|
| 66 |
+
- Datasets 3.0.0
|
| 67 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.993127147766323,
|
| 3 |
+
"total_flos": 8.781509506919629e+16,
|
| 4 |
+
"train_loss": 0.083118896693762,
|
| 5 |
+
"train_runtime": 3358.7055,
|
| 6 |
+
"train_samples": 18598,
|
| 7 |
+
"train_samples_per_second": 11.075,
|
| 8 |
+
"train_steps_per_second": 0.086
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 128000,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
128001,
|
| 6 |
+
128008,
|
| 7 |
+
128009
|
| 8 |
+
],
|
| 9 |
+
"temperature": 0.6,
|
| 10 |
+
"top_p": 0.9,
|
| 11 |
+
"transformers_version": "4.45.0.dev0"
|
| 12 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.993127147766323,
|
| 3 |
+
"total_flos": 8.781509506919629e+16,
|
| 4 |
+
"train_loss": 0.083118896693762,
|
| 5 |
+
"train_runtime": 3358.7055,
|
| 6 |
+
"train_samples": 18598,
|
| 7 |
+
"train_samples_per_second": 11.075,
|
| 8 |
+
"train_steps_per_second": 0.086
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2088 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.993127147766323,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 290,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.006872852233676976,
|
| 13 |
+
"grad_norm": 13.343527845740242,
|
| 14 |
+
"learning_rate": 3.4482758620689656e-07,
|
| 15 |
+
"loss": 0.3093,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.013745704467353952,
|
| 20 |
+
"grad_norm": 12.24978979790274,
|
| 21 |
+
"learning_rate": 6.896551724137931e-07,
|
| 22 |
+
"loss": 0.3119,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.020618556701030927,
|
| 27 |
+
"grad_norm": 11.514707536352253,
|
| 28 |
+
"learning_rate": 1.0344827586206898e-06,
|
| 29 |
+
"loss": 0.2719,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.027491408934707903,
|
| 34 |
+
"grad_norm": 10.931603411343271,
|
| 35 |
+
"learning_rate": 1.3793103448275862e-06,
|
| 36 |
+
"loss": 0.334,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.03436426116838488,
|
| 41 |
+
"grad_norm": 10.285227060724104,
|
| 42 |
+
"learning_rate": 1.724137931034483e-06,
|
| 43 |
+
"loss": 0.2734,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.041237113402061855,
|
| 48 |
+
"grad_norm": 8.121285393972324,
|
| 49 |
+
"learning_rate": 2.0689655172413796e-06,
|
| 50 |
+
"loss": 0.2204,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.048109965635738834,
|
| 55 |
+
"grad_norm": 4.851150465742376,
|
| 56 |
+
"learning_rate": 2.4137931034482762e-06,
|
| 57 |
+
"loss": 0.1903,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.054982817869415807,
|
| 62 |
+
"grad_norm": 5.611799968919774,
|
| 63 |
+
"learning_rate": 2.7586206896551725e-06,
|
| 64 |
+
"loss": 0.2094,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.061855670103092786,
|
| 69 |
+
"grad_norm": 4.177179899108234,
|
| 70 |
+
"learning_rate": 3.103448275862069e-06,
|
| 71 |
+
"loss": 0.1949,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.06872852233676977,
|
| 76 |
+
"grad_norm": 3.387529147874442,
|
| 77 |
+
"learning_rate": 3.448275862068966e-06,
|
| 78 |
+
"loss": 0.146,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.07560137457044673,
|
| 83 |
+
"grad_norm": 2.5661620699173024,
|
| 84 |
+
"learning_rate": 3.793103448275862e-06,
|
| 85 |
+
"loss": 0.1741,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.08247422680412371,
|
| 90 |
+
"grad_norm": 7.558143822067422,
|
| 91 |
+
"learning_rate": 4.137931034482759e-06,
|
| 92 |
+
"loss": 0.1945,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.08934707903780069,
|
| 97 |
+
"grad_norm": 3.003602416429719,
|
| 98 |
+
"learning_rate": 4.482758620689656e-06,
|
| 99 |
+
"loss": 0.1444,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.09621993127147767,
|
| 104 |
+
"grad_norm": 3.937942345401438,
|
| 105 |
+
"learning_rate": 4.8275862068965525e-06,
|
| 106 |
+
"loss": 0.1832,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.10309278350515463,
|
| 111 |
+
"grad_norm": 2.553710179954249,
|
| 112 |
+
"learning_rate": 5.172413793103449e-06,
|
| 113 |
+
"loss": 0.1436,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.10996563573883161,
|
| 118 |
+
"grad_norm": 3.704435645593398,
|
| 119 |
+
"learning_rate": 5.517241379310345e-06,
|
| 120 |
+
"loss": 0.1299,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.11683848797250859,
|
| 125 |
+
"grad_norm": 2.556984711627345,
|
| 126 |
+
"learning_rate": 5.862068965517242e-06,
|
| 127 |
+
"loss": 0.1472,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.12371134020618557,
|
| 132 |
+
"grad_norm": 2.6324666619403385,
|
| 133 |
+
"learning_rate": 6.206896551724138e-06,
|
| 134 |
+
"loss": 0.1293,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.13058419243986255,
|
| 139 |
+
"grad_norm": 2.0190151143609905,
|
| 140 |
+
"learning_rate": 6.551724137931035e-06,
|
| 141 |
+
"loss": 0.0939,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.13745704467353953,
|
| 146 |
+
"grad_norm": 3.000083160972002,
|
| 147 |
+
"learning_rate": 6.896551724137932e-06,
|
| 148 |
+
"loss": 0.1168,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.14432989690721648,
|
| 153 |
+
"grad_norm": 1.791742979569004,
|
| 154 |
+
"learning_rate": 7.241379310344828e-06,
|
| 155 |
+
"loss": 0.1403,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.15120274914089346,
|
| 160 |
+
"grad_norm": 1.9811274505355223,
|
| 161 |
+
"learning_rate": 7.586206896551724e-06,
|
| 162 |
+
"loss": 0.1182,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.15807560137457044,
|
| 167 |
+
"grad_norm": 2.8381777274976985,
|
| 168 |
+
"learning_rate": 7.93103448275862e-06,
|
| 169 |
+
"loss": 0.1304,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.16494845360824742,
|
| 174 |
+
"grad_norm": 1.7658433111673677,
|
| 175 |
+
"learning_rate": 8.275862068965518e-06,
|
| 176 |
+
"loss": 0.1011,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.1718213058419244,
|
| 181 |
+
"grad_norm": 2.051609499818478,
|
| 182 |
+
"learning_rate": 8.620689655172414e-06,
|
| 183 |
+
"loss": 0.1033,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.17869415807560138,
|
| 188 |
+
"grad_norm": 1.6906156790738576,
|
| 189 |
+
"learning_rate": 8.965517241379312e-06,
|
| 190 |
+
"loss": 0.0876,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.18556701030927836,
|
| 195 |
+
"grad_norm": 1.3499271620310584,
|
| 196 |
+
"learning_rate": 9.310344827586207e-06,
|
| 197 |
+
"loss": 0.1018,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.19243986254295534,
|
| 202 |
+
"grad_norm": 1.480127705709285,
|
| 203 |
+
"learning_rate": 9.655172413793105e-06,
|
| 204 |
+
"loss": 0.114,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.19931271477663232,
|
| 209 |
+
"grad_norm": 2.0713348602906327,
|
| 210 |
+
"learning_rate": 1e-05,
|
| 211 |
+
"loss": 0.1032,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.20618556701030927,
|
| 216 |
+
"grad_norm": 2.4674860694369567,
|
| 217 |
+
"learning_rate": 9.999637795788383e-06,
|
| 218 |
+
"loss": 0.1372,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.21305841924398625,
|
| 223 |
+
"grad_norm": 2.3016554549550836,
|
| 224 |
+
"learning_rate": 9.99855123563029e-06,
|
| 225 |
+
"loss": 0.0999,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.21993127147766323,
|
| 230 |
+
"grad_norm": 1.718509009864066,
|
| 231 |
+
"learning_rate": 9.996740476948386e-06,
|
| 232 |
+
"loss": 0.1185,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.2268041237113402,
|
| 237 |
+
"grad_norm": 2.861483161187408,
|
| 238 |
+
"learning_rate": 9.994205782088438e-06,
|
| 239 |
+
"loss": 0.1227,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.23367697594501718,
|
| 244 |
+
"grad_norm": 2.725246699209764,
|
| 245 |
+
"learning_rate": 9.990947518281312e-06,
|
| 246 |
+
"loss": 0.1265,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.24054982817869416,
|
| 251 |
+
"grad_norm": 1.7486506092226466,
|
| 252 |
+
"learning_rate": 9.986966157589751e-06,
|
| 253 |
+
"loss": 0.0988,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.24742268041237114,
|
| 258 |
+
"grad_norm": 2.1808324098494154,
|
| 259 |
+
"learning_rate": 9.982262276840002e-06,
|
| 260 |
+
"loss": 0.1133,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.2542955326460481,
|
| 265 |
+
"grad_norm": 2.4541258860069735,
|
| 266 |
+
"learning_rate": 9.976836557538234e-06,
|
| 267 |
+
"loss": 0.1427,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.2611683848797251,
|
| 272 |
+
"grad_norm": 1.916906089523192,
|
| 273 |
+
"learning_rate": 9.970689785771798e-06,
|
| 274 |
+
"loss": 0.1155,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.26804123711340205,
|
| 279 |
+
"grad_norm": 1.2482216180682295,
|
| 280 |
+
"learning_rate": 9.963822852095344e-06,
|
| 281 |
+
"loss": 0.0878,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.27491408934707906,
|
| 286 |
+
"grad_norm": 1.5308178734097952,
|
| 287 |
+
"learning_rate": 9.95623675140179e-06,
|
| 288 |
+
"loss": 0.1112,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.281786941580756,
|
| 293 |
+
"grad_norm": 2.0932610150905653,
|
| 294 |
+
"learning_rate": 9.947932582778188e-06,
|
| 295 |
+
"loss": 0.1226,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.28865979381443296,
|
| 300 |
+
"grad_norm": 1.7335203315950243,
|
| 301 |
+
"learning_rate": 9.938911549346473e-06,
|
| 302 |
+
"loss": 0.1046,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.29553264604810997,
|
| 307 |
+
"grad_norm": 1.3019830512685613,
|
| 308 |
+
"learning_rate": 9.929174958089167e-06,
|
| 309 |
+
"loss": 0.0732,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.3024054982817869,
|
| 314 |
+
"grad_norm": 2.173050217291364,
|
| 315 |
+
"learning_rate": 9.918724219660013e-06,
|
| 316 |
+
"loss": 0.1162,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.30927835051546393,
|
| 321 |
+
"grad_norm": 1.8878314610669742,
|
| 322 |
+
"learning_rate": 9.907560848179607e-06,
|
| 323 |
+
"loss": 0.1176,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.3161512027491409,
|
| 328 |
+
"grad_norm": 1.239886305629569,
|
| 329 |
+
"learning_rate": 9.895686461016007e-06,
|
| 330 |
+
"loss": 0.0933,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.3230240549828179,
|
| 335 |
+
"grad_norm": 1.8458615143849681,
|
| 336 |
+
"learning_rate": 9.883102778550434e-06,
|
| 337 |
+
"loss": 0.1015,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.32989690721649484,
|
| 342 |
+
"grad_norm": 0.9803248656424836,
|
| 343 |
+
"learning_rate": 9.869811623928001e-06,
|
| 344 |
+
"loss": 0.0826,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.33676975945017185,
|
| 349 |
+
"grad_norm": 1.1676847254145442,
|
| 350 |
+
"learning_rate": 9.855814922793583e-06,
|
| 351 |
+
"loss": 0.0882,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.3436426116838488,
|
| 356 |
+
"grad_norm": 1.0989486969646112,
|
| 357 |
+
"learning_rate": 9.841114703012817e-06,
|
| 358 |
+
"loss": 0.0734,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.35051546391752575,
|
| 363 |
+
"grad_norm": 1.3391376973027571,
|
| 364 |
+
"learning_rate": 9.82571309437831e-06,
|
| 365 |
+
"loss": 0.1067,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.35738831615120276,
|
| 370 |
+
"grad_norm": 1.925598063620997,
|
| 371 |
+
"learning_rate": 9.809612328301071e-06,
|
| 372 |
+
"loss": 0.1292,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.3642611683848797,
|
| 377 |
+
"grad_norm": 1.0411337983613875,
|
| 378 |
+
"learning_rate": 9.792814737487207e-06,
|
| 379 |
+
"loss": 0.0826,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.3711340206185567,
|
| 384 |
+
"grad_norm": 1.396102272741454,
|
| 385 |
+
"learning_rate": 9.775322755599979e-06,
|
| 386 |
+
"loss": 0.1306,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.37800687285223367,
|
| 391 |
+
"grad_norm": 0.7722547955211074,
|
| 392 |
+
"learning_rate": 9.757138916907184e-06,
|
| 393 |
+
"loss": 0.0768,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.3848797250859107,
|
| 398 |
+
"grad_norm": 1.0518721091861032,
|
| 399 |
+
"learning_rate": 9.738265855914014e-06,
|
| 400 |
+
"loss": 0.0866,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.3917525773195876,
|
| 405 |
+
"grad_norm": 1.0004771317087249,
|
| 406 |
+
"learning_rate": 9.718706306981332e-06,
|
| 407 |
+
"loss": 0.0857,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.39862542955326463,
|
| 412 |
+
"grad_norm": 1.2108667681479588,
|
| 413 |
+
"learning_rate": 9.698463103929542e-06,
|
| 414 |
+
"loss": 0.0872,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.4054982817869416,
|
| 419 |
+
"grad_norm": 1.0054960437725995,
|
| 420 |
+
"learning_rate": 9.677539179628005e-06,
|
| 421 |
+
"loss": 0.1149,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.41237113402061853,
|
| 426 |
+
"grad_norm": 1.3138211548976666,
|
| 427 |
+
"learning_rate": 9.655937565570124e-06,
|
| 428 |
+
"loss": 0.0963,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.41924398625429554,
|
| 433 |
+
"grad_norm": 1.1343794503691726,
|
| 434 |
+
"learning_rate": 9.63366139143413e-06,
|
| 435 |
+
"loss": 0.0974,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.4261168384879725,
|
| 440 |
+
"grad_norm": 2.639452170631618,
|
| 441 |
+
"learning_rate": 9.610713884629667e-06,
|
| 442 |
+
"loss": 0.1444,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.4329896907216495,
|
| 447 |
+
"grad_norm": 1.6733576219511557,
|
| 448 |
+
"learning_rate": 9.587098369830171e-06,
|
| 449 |
+
"loss": 0.0952,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.43986254295532645,
|
| 454 |
+
"grad_norm": 0.9798445270256229,
|
| 455 |
+
"learning_rate": 9.562818268491216e-06,
|
| 456 |
+
"loss": 0.0914,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.44673539518900346,
|
| 461 |
+
"grad_norm": 1.5699674372191357,
|
| 462 |
+
"learning_rate": 9.537877098354787e-06,
|
| 463 |
+
"loss": 0.1028,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.4536082474226804,
|
| 468 |
+
"grad_norm": 1.1340293309823666,
|
| 469 |
+
"learning_rate": 9.512278472939627e-06,
|
| 470 |
+
"loss": 0.0722,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.46048109965635736,
|
| 475 |
+
"grad_norm": 1.173682861226043,
|
| 476 |
+
"learning_rate": 9.486026101017711e-06,
|
| 477 |
+
"loss": 0.0748,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.46735395189003437,
|
| 482 |
+
"grad_norm": 1.157619190820778,
|
| 483 |
+
"learning_rate": 9.459123786076911e-06,
|
| 484 |
+
"loss": 0.0958,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.4742268041237113,
|
| 489 |
+
"grad_norm": 1.2139013816286683,
|
| 490 |
+
"learning_rate": 9.431575425769938e-06,
|
| 491 |
+
"loss": 0.0767,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.48109965635738833,
|
| 496 |
+
"grad_norm": 1.366755801620074,
|
| 497 |
+
"learning_rate": 9.40338501134964e-06,
|
| 498 |
+
"loss": 0.1277,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.4879725085910653,
|
| 503 |
+
"grad_norm": 1.698261916010506,
|
| 504 |
+
"learning_rate": 9.374556627090749e-06,
|
| 505 |
+
"loss": 0.0871,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.4948453608247423,
|
| 510 |
+
"grad_norm": 1.600045772348992,
|
| 511 |
+
"learning_rate": 9.345094449698143e-06,
|
| 512 |
+
"loss": 0.0944,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.5017182130584192,
|
| 517 |
+
"grad_norm": 1.3062219524861596,
|
| 518 |
+
"learning_rate": 9.315002747701716e-06,
|
| 519 |
+
"loss": 0.0844,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.5085910652920962,
|
| 524 |
+
"grad_norm": 1.3506971841269695,
|
| 525 |
+
"learning_rate": 9.284285880837947e-06,
|
| 526 |
+
"loss": 0.0734,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.5154639175257731,
|
| 531 |
+
"grad_norm": 1.772082990992334,
|
| 532 |
+
"learning_rate": 9.252948299418255e-06,
|
| 533 |
+
"loss": 0.0896,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.5223367697594502,
|
| 538 |
+
"grad_norm": 1.857624623179696,
|
| 539 |
+
"learning_rate": 9.220994543684225e-06,
|
| 540 |
+
"loss": 0.0987,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.5292096219931272,
|
| 545 |
+
"grad_norm": 1.2542633933357743,
|
| 546 |
+
"learning_rate": 9.188429243149824e-06,
|
| 547 |
+
"loss": 0.1108,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.5360824742268041,
|
| 552 |
+
"grad_norm": 0.7154348047087423,
|
| 553 |
+
"learning_rate": 9.155257115930651e-06,
|
| 554 |
+
"loss": 0.0743,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.5429553264604811,
|
| 559 |
+
"grad_norm": 1.1412836758516878,
|
| 560 |
+
"learning_rate": 9.121482968060384e-06,
|
| 561 |
+
"loss": 0.0819,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.5498281786941581,
|
| 566 |
+
"grad_norm": 1.3046778978500768,
|
| 567 |
+
"learning_rate": 9.08711169279446e-06,
|
| 568 |
+
"loss": 0.1218,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.5567010309278351,
|
| 573 |
+
"grad_norm": 0.9587676579566864,
|
| 574 |
+
"learning_rate": 9.052148269901145e-06,
|
| 575 |
+
"loss": 0.0756,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.563573883161512,
|
| 580 |
+
"grad_norm": 1.28525105455312,
|
| 581 |
+
"learning_rate": 9.01659776494005e-06,
|
| 582 |
+
"loss": 0.084,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.570446735395189,
|
| 587 |
+
"grad_norm": 1.4574786423091366,
|
| 588 |
+
"learning_rate": 8.98046532852822e-06,
|
| 589 |
+
"loss": 0.1231,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.5773195876288659,
|
| 594 |
+
"grad_norm": 0.7561541721320211,
|
| 595 |
+
"learning_rate": 8.943756195593916e-06,
|
| 596 |
+
"loss": 0.0668,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.584192439862543,
|
| 601 |
+
"grad_norm": 0.8272606217406574,
|
| 602 |
+
"learning_rate": 8.90647568461816e-06,
|
| 603 |
+
"loss": 0.0724,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.5910652920962199,
|
| 608 |
+
"grad_norm": 1.1658084746326034,
|
| 609 |
+
"learning_rate": 8.868629196864182e-06,
|
| 610 |
+
"loss": 0.0988,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.5979381443298969,
|
| 615 |
+
"grad_norm": 1.0267861508467409,
|
| 616 |
+
"learning_rate": 8.83022221559489e-06,
|
| 617 |
+
"loss": 0.1028,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.6048109965635738,
|
| 622 |
+
"grad_norm": 1.048307624038121,
|
| 623 |
+
"learning_rate": 8.791260305278434e-06,
|
| 624 |
+
"loss": 0.0779,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.6116838487972509,
|
| 629 |
+
"grad_norm": 1.3665792411125233,
|
| 630 |
+
"learning_rate": 8.751749110782013e-06,
|
| 631 |
+
"loss": 0.0987,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.6185567010309279,
|
| 636 |
+
"grad_norm": 0.9932979614965134,
|
| 637 |
+
"learning_rate": 8.71169435655405e-06,
|
| 638 |
+
"loss": 0.0759,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.6254295532646048,
|
| 643 |
+
"grad_norm": 1.1150239614480977,
|
| 644 |
+
"learning_rate": 8.671101845794816e-06,
|
| 645 |
+
"loss": 0.0975,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.6323024054982818,
|
| 650 |
+
"grad_norm": 1.222073797031231,
|
| 651 |
+
"learning_rate": 8.629977459615655e-06,
|
| 652 |
+
"loss": 0.1083,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.6391752577319587,
|
| 657 |
+
"grad_norm": 0.7113417636139049,
|
| 658 |
+
"learning_rate": 8.588327156186915e-06,
|
| 659 |
+
"loss": 0.0645,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.6460481099656358,
|
| 664 |
+
"grad_norm": 0.7517765627447859,
|
| 665 |
+
"learning_rate": 8.546156969874723e-06,
|
| 666 |
+
"loss": 0.0789,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.6529209621993127,
|
| 671 |
+
"grad_norm": 0.9975696135564021,
|
| 672 |
+
"learning_rate": 8.503473010366713e-06,
|
| 673 |
+
"loss": 0.0974,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.6597938144329897,
|
| 678 |
+
"grad_norm": 1.5149352410721835,
|
| 679 |
+
"learning_rate": 8.460281461786848e-06,
|
| 680 |
+
"loss": 0.1121,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.6666666666666666,
|
| 685 |
+
"grad_norm": 0.6999416336418273,
|
| 686 |
+
"learning_rate": 8.416588581799447e-06,
|
| 687 |
+
"loss": 0.0983,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.6735395189003437,
|
| 692 |
+
"grad_norm": 0.9288258345523339,
|
| 693 |
+
"learning_rate": 8.372400700702569e-06,
|
| 694 |
+
"loss": 0.0981,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.6804123711340206,
|
| 699 |
+
"grad_norm": 0.679617249303424,
|
| 700 |
+
"learning_rate": 8.327724220510873e-06,
|
| 701 |
+
"loss": 0.0749,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.6872852233676976,
|
| 706 |
+
"grad_norm": 0.9613737743860814,
|
| 707 |
+
"learning_rate": 8.282565614028068e-06,
|
| 708 |
+
"loss": 0.0839,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.6941580756013745,
|
| 713 |
+
"grad_norm": 1.233132976094516,
|
| 714 |
+
"learning_rate": 8.23693142390914e-06,
|
| 715 |
+
"loss": 0.091,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.7010309278350515,
|
| 720 |
+
"grad_norm": 0.9827076763530673,
|
| 721 |
+
"learning_rate": 8.19082826171243e-06,
|
| 722 |
+
"loss": 0.0837,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.7079037800687286,
|
| 727 |
+
"grad_norm": 1.1935581064453669,
|
| 728 |
+
"learning_rate": 8.144262806941743e-06,
|
| 729 |
+
"loss": 0.0687,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.7147766323024055,
|
| 734 |
+
"grad_norm": 0.9177112496125246,
|
| 735 |
+
"learning_rate": 8.097241806078616e-06,
|
| 736 |
+
"loss": 0.0777,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.7216494845360825,
|
| 741 |
+
"grad_norm": 1.2467643199963998,
|
| 742 |
+
"learning_rate": 8.049772071604864e-06,
|
| 743 |
+
"loss": 0.0903,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.7285223367697594,
|
| 748 |
+
"grad_norm": 0.9724536551102111,
|
| 749 |
+
"learning_rate": 8.001860481015594e-06,
|
| 750 |
+
"loss": 0.0618,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.7353951890034365,
|
| 755 |
+
"grad_norm": 0.7763240873384808,
|
| 756 |
+
"learning_rate": 7.953513975822755e-06,
|
| 757 |
+
"loss": 0.0689,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.7422680412371134,
|
| 762 |
+
"grad_norm": 1.2003193887125871,
|
| 763 |
+
"learning_rate": 7.904739560549475e-06,
|
| 764 |
+
"loss": 0.1041,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.7491408934707904,
|
| 769 |
+
"grad_norm": 0.8958459053146742,
|
| 770 |
+
"learning_rate": 7.855544301715203e-06,
|
| 771 |
+
"loss": 0.0862,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.7560137457044673,
|
| 776 |
+
"grad_norm": 1.3307827246787383,
|
| 777 |
+
"learning_rate": 7.805935326811913e-06,
|
| 778 |
+
"loss": 0.0962,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.7628865979381443,
|
| 783 |
+
"grad_norm": 0.7569593754166503,
|
| 784 |
+
"learning_rate": 7.755919823271466e-06,
|
| 785 |
+
"loss": 0.0782,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.7697594501718213,
|
| 790 |
+
"grad_norm": 1.0516563986684642,
|
| 791 |
+
"learning_rate": 7.70550503742427e-06,
|
| 792 |
+
"loss": 0.1127,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.7766323024054983,
|
| 797 |
+
"grad_norm": 1.0569807870257548,
|
| 798 |
+
"learning_rate": 7.654698273449435e-06,
|
| 799 |
+
"loss": 0.0951,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.7835051546391752,
|
| 804 |
+
"grad_norm": 1.552862312299585,
|
| 805 |
+
"learning_rate": 7.603506892316513e-06,
|
| 806 |
+
"loss": 0.1092,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.7903780068728522,
|
| 811 |
+
"grad_norm": 0.7997759741193654,
|
| 812 |
+
"learning_rate": 7.551938310719043e-06,
|
| 813 |
+
"loss": 0.0871,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.7972508591065293,
|
| 818 |
+
"grad_norm": 0.8412301095181958,
|
| 819 |
+
"learning_rate": 7.500000000000001e-06,
|
| 820 |
+
"loss": 0.0687,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.8041237113402062,
|
| 825 |
+
"grad_norm": 1.264393216361882,
|
| 826 |
+
"learning_rate": 7.447699485069342e-06,
|
| 827 |
+
"loss": 0.0912,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.8109965635738832,
|
| 832 |
+
"grad_norm": 0.9282362951638996,
|
| 833 |
+
"learning_rate": 7.395044343313777e-06,
|
| 834 |
+
"loss": 0.0913,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.8178694158075601,
|
| 839 |
+
"grad_norm": 0.7991005314806403,
|
| 840 |
+
"learning_rate": 7.342042203498952e-06,
|
| 841 |
+
"loss": 0.0674,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.8247422680412371,
|
| 846 |
+
"grad_norm": 1.2460777708028443,
|
| 847 |
+
"learning_rate": 7.288700744664167e-06,
|
| 848 |
+
"loss": 0.0787,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.8316151202749141,
|
| 853 |
+
"grad_norm": 0.7430347040542368,
|
| 854 |
+
"learning_rate": 7.235027695009846e-06,
|
| 855 |
+
"loss": 0.0688,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.8384879725085911,
|
| 860 |
+
"grad_norm": 0.8572145864262852,
|
| 861 |
+
"learning_rate": 7.181030830777838e-06,
|
| 862 |
+
"loss": 0.0798,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.845360824742268,
|
| 867 |
+
"grad_norm": 0.6167581560270669,
|
| 868 |
+
"learning_rate": 7.1267179751248005e-06,
|
| 869 |
+
"loss": 0.0474,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.852233676975945,
|
| 874 |
+
"grad_norm": 0.8924648363889269,
|
| 875 |
+
"learning_rate": 7.0720969969887595e-06,
|
| 876 |
+
"loss": 0.0772,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.8591065292096219,
|
| 881 |
+
"grad_norm": 0.9412488028180099,
|
| 882 |
+
"learning_rate": 7.017175809949044e-06,
|
| 883 |
+
"loss": 0.1008,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.865979381443299,
|
| 888 |
+
"grad_norm": 0.7861480916341242,
|
| 889 |
+
"learning_rate": 6.961962371079752e-06,
|
| 890 |
+
"loss": 0.0799,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.872852233676976,
|
| 895 |
+
"grad_norm": 0.6224897653682172,
|
| 896 |
+
"learning_rate": 6.906464679796927e-06,
|
| 897 |
+
"loss": 0.0665,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.8797250859106529,
|
| 902 |
+
"grad_norm": 1.425022497948844,
|
| 903 |
+
"learning_rate": 6.850690776699574e-06,
|
| 904 |
+
"loss": 0.0856,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.8865979381443299,
|
| 909 |
+
"grad_norm": 0.8333767406840923,
|
| 910 |
+
"learning_rate": 6.79464874240473e-06,
|
| 911 |
+
"loss": 0.0764,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.8934707903780069,
|
| 916 |
+
"grad_norm": 0.7721273820927107,
|
| 917 |
+
"learning_rate": 6.7383466963767386e-06,
|
| 918 |
+
"loss": 0.0699,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.9003436426116839,
|
| 923 |
+
"grad_norm": 0.8231175394948014,
|
| 924 |
+
"learning_rate": 6.681792795750876e-06,
|
| 925 |
+
"loss": 0.0832,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.9072164948453608,
|
| 930 |
+
"grad_norm": 0.6350176361674553,
|
| 931 |
+
"learning_rate": 6.624995234151539e-06,
|
| 932 |
+
"loss": 0.0638,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.9140893470790378,
|
| 937 |
+
"grad_norm": 1.370935237548587,
|
| 938 |
+
"learning_rate": 6.567962240505136e-06,
|
| 939 |
+
"loss": 0.0898,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.9209621993127147,
|
| 944 |
+
"grad_norm": 1.0895618461397043,
|
| 945 |
+
"learning_rate": 6.510702077847864e-06,
|
| 946 |
+
"loss": 0.0774,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.9278350515463918,
|
| 951 |
+
"grad_norm": 0.8146992842174668,
|
| 952 |
+
"learning_rate": 6.453223042128556e-06,
|
| 953 |
+
"loss": 0.0706,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.9347079037800687,
|
| 958 |
+
"grad_norm": 1.1421370223704364,
|
| 959 |
+
"learning_rate": 6.395533461006736e-06,
|
| 960 |
+
"loss": 0.0856,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.9415807560137457,
|
| 965 |
+
"grad_norm": 0.5899868445308769,
|
| 966 |
+
"learning_rate": 6.337641692646106e-06,
|
| 967 |
+
"loss": 0.063,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.9484536082474226,
|
| 972 |
+
"grad_norm": 1.0080688723541746,
|
| 973 |
+
"learning_rate": 6.2795561245035895e-06,
|
| 974 |
+
"loss": 0.091,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.9553264604810997,
|
| 979 |
+
"grad_norm": 0.6152604803574195,
|
| 980 |
+
"learning_rate": 6.221285172114156e-06,
|
| 981 |
+
"loss": 0.0516,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.9621993127147767,
|
| 986 |
+
"grad_norm": 0.915869955227986,
|
| 987 |
+
"learning_rate": 6.162837277871553e-06,
|
| 988 |
+
"loss": 0.0685,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.9690721649484536,
|
| 993 |
+
"grad_norm": 0.6556625739889971,
|
| 994 |
+
"learning_rate": 6.104220909805162e-06,
|
| 995 |
+
"loss": 0.0751,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.9759450171821306,
|
| 1000 |
+
"grad_norm": 0.8690225384999679,
|
| 1001 |
+
"learning_rate": 6.045444560353136e-06,
|
| 1002 |
+
"loss": 0.0801,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.9828178694158075,
|
| 1007 |
+
"grad_norm": 1.236768016408407,
|
| 1008 |
+
"learning_rate": 5.986516745132e-06,
|
| 1009 |
+
"loss": 0.0931,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.9896907216494846,
|
| 1014 |
+
"grad_norm": 0.646511655105778,
|
| 1015 |
+
"learning_rate": 5.927446001702899e-06,
|
| 1016 |
+
"loss": 0.0633,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.9965635738831615,
|
| 1021 |
+
"grad_norm": 0.931895984228249,
|
| 1022 |
+
"learning_rate": 5.8682408883346535e-06,
|
| 1023 |
+
"loss": 0.0963,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.9965635738831615,
|
| 1028 |
+
"eval_loss": 0.07353807240724564,
|
| 1029 |
+
"eval_runtime": 28.6691,
|
| 1030 |
+
"eval_samples_per_second": 34.148,
|
| 1031 |
+
"eval_steps_per_second": 1.081,
|
| 1032 |
+
"step": 145
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.0034364261168385,
|
| 1036 |
+
"grad_norm": 0.6089204452972106,
|
| 1037 |
+
"learning_rate": 5.808909982763825e-06,
|
| 1038 |
+
"loss": 0.0592,
|
| 1039 |
+
"step": 146
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.0103092783505154,
|
| 1043 |
+
"grad_norm": 0.711364999370767,
|
| 1044 |
+
"learning_rate": 5.749461880951966e-06,
|
| 1045 |
+
"loss": 0.0728,
|
| 1046 |
+
"step": 147
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.0171821305841924,
|
| 1050 |
+
"grad_norm": 0.8200567189598971,
|
| 1051 |
+
"learning_rate": 5.689905195840216e-06,
|
| 1052 |
+
"loss": 0.0439,
|
| 1053 |
+
"step": 148
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.0240549828178693,
|
| 1057 |
+
"grad_norm": 0.822148606255475,
|
| 1058 |
+
"learning_rate": 5.630248556101448e-06,
|
| 1059 |
+
"loss": 0.0672,
|
| 1060 |
+
"step": 149
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.0309278350515463,
|
| 1064 |
+
"grad_norm": 0.9665507477611337,
|
| 1065 |
+
"learning_rate": 5.570500604890124e-06,
|
| 1066 |
+
"loss": 0.0798,
|
| 1067 |
+
"step": 150
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.0378006872852235,
|
| 1071 |
+
"grad_norm": 0.5435621364714303,
|
| 1072 |
+
"learning_rate": 5.510669998590074e-06,
|
| 1073 |
+
"loss": 0.0409,
|
| 1074 |
+
"step": 151
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.0446735395189004,
|
| 1078 |
+
"grad_norm": 0.6688474665139779,
|
| 1079 |
+
"learning_rate": 5.450765405560328e-06,
|
| 1080 |
+
"loss": 0.061,
|
| 1081 |
+
"step": 152
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.0515463917525774,
|
| 1085 |
+
"grad_norm": 0.8667629627483111,
|
| 1086 |
+
"learning_rate": 5.390795504879243e-06,
|
| 1087 |
+
"loss": 0.062,
|
| 1088 |
+
"step": 153
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.0584192439862543,
|
| 1092 |
+
"grad_norm": 0.8748519249664792,
|
| 1093 |
+
"learning_rate": 5.330768985087059e-06,
|
| 1094 |
+
"loss": 0.0736,
|
| 1095 |
+
"step": 154
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.0652920962199313,
|
| 1099 |
+
"grad_norm": 0.8459919910817141,
|
| 1100 |
+
"learning_rate": 5.270694542927089e-06,
|
| 1101 |
+
"loss": 0.1011,
|
| 1102 |
+
"step": 155
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.0721649484536082,
|
| 1106 |
+
"grad_norm": 1.0006807281026835,
|
| 1107 |
+
"learning_rate": 5.2105808820857126e-06,
|
| 1108 |
+
"loss": 0.0754,
|
| 1109 |
+
"step": 156
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.0790378006872852,
|
| 1113 |
+
"grad_norm": 0.6893114221570908,
|
| 1114 |
+
"learning_rate": 5.150436711931387e-06,
|
| 1115 |
+
"loss": 0.0636,
|
| 1116 |
+
"step": 157
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.0859106529209621,
|
| 1120 |
+
"grad_norm": 0.6784930300682643,
|
| 1121 |
+
"learning_rate": 5.090270746252803e-06,
|
| 1122 |
+
"loss": 0.0601,
|
| 1123 |
+
"step": 158
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.0927835051546393,
|
| 1127 |
+
"grad_norm": 0.5114841540763814,
|
| 1128 |
+
"learning_rate": 5.030091701996428e-06,
|
| 1129 |
+
"loss": 0.0497,
|
| 1130 |
+
"step": 159
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.0996563573883162,
|
| 1134 |
+
"grad_norm": 0.866460034783626,
|
| 1135 |
+
"learning_rate": 4.9699082980035735e-06,
|
| 1136 |
+
"loss": 0.0664,
|
| 1137 |
+
"step": 160
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.1065292096219932,
|
| 1141 |
+
"grad_norm": 0.8224798726291673,
|
| 1142 |
+
"learning_rate": 4.909729253747197e-06,
|
| 1143 |
+
"loss": 0.0684,
|
| 1144 |
+
"step": 161
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.1134020618556701,
|
| 1148 |
+
"grad_norm": 0.7028489571924831,
|
| 1149 |
+
"learning_rate": 4.8495632880686155e-06,
|
| 1150 |
+
"loss": 0.0552,
|
| 1151 |
+
"step": 162
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.120274914089347,
|
| 1155 |
+
"grad_norm": 0.6830461960827369,
|
| 1156 |
+
"learning_rate": 4.789419117914288e-06,
|
| 1157 |
+
"loss": 0.0544,
|
| 1158 |
+
"step": 163
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.127147766323024,
|
| 1162 |
+
"grad_norm": 0.49669647904528813,
|
| 1163 |
+
"learning_rate": 4.729305457072913e-06,
|
| 1164 |
+
"loss": 0.0472,
|
| 1165 |
+
"step": 164
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.134020618556701,
|
| 1169 |
+
"grad_norm": 0.5504034342644836,
|
| 1170 |
+
"learning_rate": 4.6692310149129425e-06,
|
| 1171 |
+
"loss": 0.038,
|
| 1172 |
+
"step": 165
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.140893470790378,
|
| 1176 |
+
"grad_norm": 1.3003587000347374,
|
| 1177 |
+
"learning_rate": 4.609204495120759e-06,
|
| 1178 |
+
"loss": 0.0756,
|
| 1179 |
+
"step": 166
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.147766323024055,
|
| 1183 |
+
"grad_norm": 0.6139063543812373,
|
| 1184 |
+
"learning_rate": 4.549234594439674e-06,
|
| 1185 |
+
"loss": 0.0462,
|
| 1186 |
+
"step": 167
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.1546391752577319,
|
| 1190 |
+
"grad_norm": 0.8970591982798499,
|
| 1191 |
+
"learning_rate": 4.489330001409929e-06,
|
| 1192 |
+
"loss": 0.0614,
|
| 1193 |
+
"step": 168
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.161512027491409,
|
| 1197 |
+
"grad_norm": 0.5995639690476058,
|
| 1198 |
+
"learning_rate": 4.429499395109877e-06,
|
| 1199 |
+
"loss": 0.0632,
|
| 1200 |
+
"step": 169
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.168384879725086,
|
| 1204 |
+
"grad_norm": 0.6529977139603435,
|
| 1205 |
+
"learning_rate": 4.369751443898554e-06,
|
| 1206 |
+
"loss": 0.0711,
|
| 1207 |
+
"step": 170
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.175257731958763,
|
| 1211 |
+
"grad_norm": 0.8174312432811807,
|
| 1212 |
+
"learning_rate": 4.310094804159784e-06,
|
| 1213 |
+
"loss": 0.0715,
|
| 1214 |
+
"step": 171
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.1821305841924399,
|
| 1218 |
+
"grad_norm": 0.47250078421627173,
|
| 1219 |
+
"learning_rate": 4.250538119048036e-06,
|
| 1220 |
+
"loss": 0.057,
|
| 1221 |
+
"step": 172
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.1890034364261168,
|
| 1225 |
+
"grad_norm": 0.5565834381318029,
|
| 1226 |
+
"learning_rate": 4.191090017236177e-06,
|
| 1227 |
+
"loss": 0.0531,
|
| 1228 |
+
"step": 173
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.1958762886597938,
|
| 1232 |
+
"grad_norm": 0.5896646155456795,
|
| 1233 |
+
"learning_rate": 4.131759111665349e-06,
|
| 1234 |
+
"loss": 0.0586,
|
| 1235 |
+
"step": 174
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.2027491408934707,
|
| 1239 |
+
"grad_norm": 0.5418688448155202,
|
| 1240 |
+
"learning_rate": 4.072553998297103e-06,
|
| 1241 |
+
"loss": 0.0591,
|
| 1242 |
+
"step": 175
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.2096219931271477,
|
| 1246 |
+
"grad_norm": 0.7328302470539767,
|
| 1247 |
+
"learning_rate": 4.013483254868001e-06,
|
| 1248 |
+
"loss": 0.0765,
|
| 1249 |
+
"step": 176
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.2164948453608249,
|
| 1253 |
+
"grad_norm": 0.8882765824885585,
|
| 1254 |
+
"learning_rate": 3.9545554396468655e-06,
|
| 1255 |
+
"loss": 0.1069,
|
| 1256 |
+
"step": 177
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.2233676975945018,
|
| 1260 |
+
"grad_norm": 0.5211368359978594,
|
| 1261 |
+
"learning_rate": 3.89577909019484e-06,
|
| 1262 |
+
"loss": 0.0593,
|
| 1263 |
+
"step": 178
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.2302405498281788,
|
| 1267 |
+
"grad_norm": 0.5739804738679034,
|
| 1268 |
+
"learning_rate": 3.8371627221284495e-06,
|
| 1269 |
+
"loss": 0.0516,
|
| 1270 |
+
"step": 179
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.2371134020618557,
|
| 1274 |
+
"grad_norm": 0.7063296459732961,
|
| 1275 |
+
"learning_rate": 3.7787148278858453e-06,
|
| 1276 |
+
"loss": 0.0774,
|
| 1277 |
+
"step": 180
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.2439862542955327,
|
| 1281 |
+
"grad_norm": 0.6246259366658415,
|
| 1282 |
+
"learning_rate": 3.7204438754964113e-06,
|
| 1283 |
+
"loss": 0.0668,
|
| 1284 |
+
"step": 181
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.2508591065292096,
|
| 1288 |
+
"grad_norm": 0.7085735957910768,
|
| 1289 |
+
"learning_rate": 3.662358307353897e-06,
|
| 1290 |
+
"loss": 0.0659,
|
| 1291 |
+
"step": 182
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.2577319587628866,
|
| 1295 |
+
"grad_norm": 0.6498297940131951,
|
| 1296 |
+
"learning_rate": 3.6044665389932663e-06,
|
| 1297 |
+
"loss": 0.0775,
|
| 1298 |
+
"step": 183
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.2646048109965635,
|
| 1302 |
+
"grad_norm": 0.5412566210037862,
|
| 1303 |
+
"learning_rate": 3.5467769578714455e-06,
|
| 1304 |
+
"loss": 0.0589,
|
| 1305 |
+
"step": 184
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.2714776632302405,
|
| 1309 |
+
"grad_norm": 0.648725736061613,
|
| 1310 |
+
"learning_rate": 3.489297922152136e-06,
|
| 1311 |
+
"loss": 0.056,
|
| 1312 |
+
"step": 185
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.2783505154639174,
|
| 1316 |
+
"grad_norm": 0.6779616929655204,
|
| 1317 |
+
"learning_rate": 3.432037759494867e-06,
|
| 1318 |
+
"loss": 0.0553,
|
| 1319 |
+
"step": 186
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.2852233676975944,
|
| 1323 |
+
"grad_norm": 0.6253730030455485,
|
| 1324 |
+
"learning_rate": 3.375004765848463e-06,
|
| 1325 |
+
"loss": 0.0514,
|
| 1326 |
+
"step": 187
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.2920962199312716,
|
| 1330 |
+
"grad_norm": 0.9750836461661787,
|
| 1331 |
+
"learning_rate": 3.3182072042491244e-06,
|
| 1332 |
+
"loss": 0.0741,
|
| 1333 |
+
"step": 188
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.2989690721649485,
|
| 1337 |
+
"grad_norm": 0.7297987870924404,
|
| 1338 |
+
"learning_rate": 3.2616533036232635e-06,
|
| 1339 |
+
"loss": 0.0555,
|
| 1340 |
+
"step": 189
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.3058419243986255,
|
| 1344 |
+
"grad_norm": 0.7949515783989405,
|
| 1345 |
+
"learning_rate": 3.205351257595272e-06,
|
| 1346 |
+
"loss": 0.0701,
|
| 1347 |
+
"step": 190
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.3127147766323024,
|
| 1351 |
+
"grad_norm": 0.691303406094663,
|
| 1352 |
+
"learning_rate": 3.149309223300428e-06,
|
| 1353 |
+
"loss": 0.0522,
|
| 1354 |
+
"step": 191
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.3195876288659794,
|
| 1358 |
+
"grad_norm": 0.752802739333483,
|
| 1359 |
+
"learning_rate": 3.093535320203074e-06,
|
| 1360 |
+
"loss": 0.0632,
|
| 1361 |
+
"step": 192
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.3264604810996563,
|
| 1365 |
+
"grad_norm": 0.5981220801012371,
|
| 1366 |
+
"learning_rate": 3.0380376289202497e-06,
|
| 1367 |
+
"loss": 0.0502,
|
| 1368 |
+
"step": 193
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.3333333333333333,
|
| 1372 |
+
"grad_norm": 0.6416551923363477,
|
| 1373 |
+
"learning_rate": 2.982824190050958e-06,
|
| 1374 |
+
"loss": 0.0601,
|
| 1375 |
+
"step": 194
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.3402061855670104,
|
| 1379 |
+
"grad_norm": 0.677260132302546,
|
| 1380 |
+
"learning_rate": 2.927903003011241e-06,
|
| 1381 |
+
"loss": 0.0621,
|
| 1382 |
+
"step": 195
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.3470790378006874,
|
| 1386 |
+
"grad_norm": 0.8395402924477804,
|
| 1387 |
+
"learning_rate": 2.8732820248752016e-06,
|
| 1388 |
+
"loss": 0.048,
|
| 1389 |
+
"step": 196
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.3539518900343643,
|
| 1393 |
+
"grad_norm": 0.8002606635988642,
|
| 1394 |
+
"learning_rate": 2.8189691692221627e-06,
|
| 1395 |
+
"loss": 0.0834,
|
| 1396 |
+
"step": 197
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.3608247422680413,
|
| 1400 |
+
"grad_norm": 0.5077816097736637,
|
| 1401 |
+
"learning_rate": 2.7649723049901554e-06,
|
| 1402 |
+
"loss": 0.0583,
|
| 1403 |
+
"step": 198
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.3676975945017182,
|
| 1407 |
+
"grad_norm": 0.6529932183111826,
|
| 1408 |
+
"learning_rate": 2.711299255335833e-06,
|
| 1409 |
+
"loss": 0.0647,
|
| 1410 |
+
"step": 199
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.3745704467353952,
|
| 1414 |
+
"grad_norm": 0.8015571104493192,
|
| 1415 |
+
"learning_rate": 2.65795779650105e-06,
|
| 1416 |
+
"loss": 0.0728,
|
| 1417 |
+
"step": 200
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.3814432989690721,
|
| 1421 |
+
"grad_norm": 0.71051287679919,
|
| 1422 |
+
"learning_rate": 2.6049556566862234e-06,
|
| 1423 |
+
"loss": 0.0612,
|
| 1424 |
+
"step": 201
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.388316151202749,
|
| 1428 |
+
"grad_norm": 0.6321188834283786,
|
| 1429 |
+
"learning_rate": 2.552300514930657e-06,
|
| 1430 |
+
"loss": 0.0601,
|
| 1431 |
+
"step": 202
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.395189003436426,
|
| 1435 |
+
"grad_norm": 0.6965828786151593,
|
| 1436 |
+
"learning_rate": 2.5000000000000015e-06,
|
| 1437 |
+
"loss": 0.0822,
|
| 1438 |
+
"step": 203
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.402061855670103,
|
| 1442 |
+
"grad_norm": 0.5670784012969757,
|
| 1443 |
+
"learning_rate": 2.4480616892809593e-06,
|
| 1444 |
+
"loss": 0.0468,
|
| 1445 |
+
"step": 204
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.40893470790378,
|
| 1449 |
+
"grad_norm": 0.6000138699421909,
|
| 1450 |
+
"learning_rate": 2.396493107683488e-06,
|
| 1451 |
+
"loss": 0.0529,
|
| 1452 |
+
"step": 205
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.4158075601374571,
|
| 1456 |
+
"grad_norm": 0.5398311870981869,
|
| 1457 |
+
"learning_rate": 2.345301726550567e-06,
|
| 1458 |
+
"loss": 0.0625,
|
| 1459 |
+
"step": 206
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.422680412371134,
|
| 1463 |
+
"grad_norm": 0.6411275132355324,
|
| 1464 |
+
"learning_rate": 2.2944949625757295e-06,
|
| 1465 |
+
"loss": 0.067,
|
| 1466 |
+
"step": 207
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.429553264604811,
|
| 1470 |
+
"grad_norm": 0.6414473827499029,
|
| 1471 |
+
"learning_rate": 2.244080176728536e-06,
|
| 1472 |
+
"loss": 0.0818,
|
| 1473 |
+
"step": 208
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.436426116838488,
|
| 1477 |
+
"grad_norm": 0.6076767228948714,
|
| 1478 |
+
"learning_rate": 2.1940646731880887e-06,
|
| 1479 |
+
"loss": 0.0623,
|
| 1480 |
+
"step": 209
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.443298969072165,
|
| 1484 |
+
"grad_norm": 0.606752985491122,
|
| 1485 |
+
"learning_rate": 2.1444556982847996e-06,
|
| 1486 |
+
"loss": 0.0721,
|
| 1487 |
+
"step": 210
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.4501718213058419,
|
| 1491 |
+
"grad_norm": 0.5466729444655251,
|
| 1492 |
+
"learning_rate": 2.095260439450526e-06,
|
| 1493 |
+
"loss": 0.0517,
|
| 1494 |
+
"step": 211
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.4570446735395188,
|
| 1498 |
+
"grad_norm": 0.6501972422129079,
|
| 1499 |
+
"learning_rate": 2.0464860241772454e-06,
|
| 1500 |
+
"loss": 0.0743,
|
| 1501 |
+
"step": 212
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.463917525773196,
|
| 1505 |
+
"grad_norm": 0.4942335971811572,
|
| 1506 |
+
"learning_rate": 1.998139518984409e-06,
|
| 1507 |
+
"loss": 0.0445,
|
| 1508 |
+
"step": 213
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.470790378006873,
|
| 1512 |
+
"grad_norm": 0.7614445684780939,
|
| 1513 |
+
"learning_rate": 1.9502279283951363e-06,
|
| 1514 |
+
"loss": 0.0708,
|
| 1515 |
+
"step": 214
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.47766323024055,
|
| 1519 |
+
"grad_norm": 0.523925636608075,
|
| 1520 |
+
"learning_rate": 1.9027581939213852e-06,
|
| 1521 |
+
"loss": 0.0379,
|
| 1522 |
+
"step": 215
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.4845360824742269,
|
| 1526 |
+
"grad_norm": 0.5788628859055078,
|
| 1527 |
+
"learning_rate": 1.8557371930582579e-06,
|
| 1528 |
+
"loss": 0.0588,
|
| 1529 |
+
"step": 216
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.4914089347079038,
|
| 1533 |
+
"grad_norm": 0.5852664035038997,
|
| 1534 |
+
"learning_rate": 1.8091717382875723e-06,
|
| 1535 |
+
"loss": 0.0534,
|
| 1536 |
+
"step": 217
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.4982817869415808,
|
| 1540 |
+
"grad_norm": 0.5070739101066728,
|
| 1541 |
+
"learning_rate": 1.7630685760908623e-06,
|
| 1542 |
+
"loss": 0.0591,
|
| 1543 |
+
"step": 218
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.5051546391752577,
|
| 1547 |
+
"grad_norm": 0.5165996803583804,
|
| 1548 |
+
"learning_rate": 1.7174343859719334e-06,
|
| 1549 |
+
"loss": 0.036,
|
| 1550 |
+
"step": 219
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.5120274914089347,
|
| 1554 |
+
"grad_norm": 0.8664930265850154,
|
| 1555 |
+
"learning_rate": 1.6722757794891287e-06,
|
| 1556 |
+
"loss": 0.0801,
|
| 1557 |
+
"step": 220
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.5189003436426116,
|
| 1561 |
+
"grad_norm": 0.5961343486397762,
|
| 1562 |
+
"learning_rate": 1.627599299297431e-06,
|
| 1563 |
+
"loss": 0.047,
|
| 1564 |
+
"step": 221
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.5257731958762886,
|
| 1568 |
+
"grad_norm": 0.8310684269047606,
|
| 1569 |
+
"learning_rate": 1.5834114182005544e-06,
|
| 1570 |
+
"loss": 0.0769,
|
| 1571 |
+
"step": 222
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.5326460481099655,
|
| 1575 |
+
"grad_norm": 0.6352898620800099,
|
| 1576 |
+
"learning_rate": 1.5397185382131524e-06,
|
| 1577 |
+
"loss": 0.0669,
|
| 1578 |
+
"step": 223
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.5395189003436425,
|
| 1582 |
+
"grad_norm": 0.43657341049512355,
|
| 1583 |
+
"learning_rate": 1.4965269896332884e-06,
|
| 1584 |
+
"loss": 0.0307,
|
| 1585 |
+
"step": 224
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.5463917525773194,
|
| 1589 |
+
"grad_norm": 0.9308468269514949,
|
| 1590 |
+
"learning_rate": 1.4538430301252783e-06,
|
| 1591 |
+
"loss": 0.0808,
|
| 1592 |
+
"step": 225
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.5532646048109966,
|
| 1596 |
+
"grad_norm": 0.6995089689913733,
|
| 1597 |
+
"learning_rate": 1.411672843813086e-06,
|
| 1598 |
+
"loss": 0.05,
|
| 1599 |
+
"step": 226
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.5601374570446735,
|
| 1603 |
+
"grad_norm": 0.7792179024831314,
|
| 1604 |
+
"learning_rate": 1.370022540384347e-06,
|
| 1605 |
+
"loss": 0.0715,
|
| 1606 |
+
"step": 227
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.5670103092783505,
|
| 1610 |
+
"grad_norm": 0.5735091512838495,
|
| 1611 |
+
"learning_rate": 1.3288981542051844e-06,
|
| 1612 |
+
"loss": 0.0823,
|
| 1613 |
+
"step": 228
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.5738831615120275,
|
| 1617 |
+
"grad_norm": 0.5872221566366822,
|
| 1618 |
+
"learning_rate": 1.2883056434459506e-06,
|
| 1619 |
+
"loss": 0.0583,
|
| 1620 |
+
"step": 229
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.5807560137457046,
|
| 1624 |
+
"grad_norm": 0.6275265841468954,
|
| 1625 |
+
"learning_rate": 1.2482508892179884e-06,
|
| 1626 |
+
"loss": 0.0433,
|
| 1627 |
+
"step": 230
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.5876288659793816,
|
| 1631 |
+
"grad_norm": 0.7429233129616758,
|
| 1632 |
+
"learning_rate": 1.2087396947215678e-06,
|
| 1633 |
+
"loss": 0.0564,
|
| 1634 |
+
"step": 231
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.5945017182130585,
|
| 1638 |
+
"grad_norm": 0.580860450601958,
|
| 1639 |
+
"learning_rate": 1.1697777844051105e-06,
|
| 1640 |
+
"loss": 0.0511,
|
| 1641 |
+
"step": 232
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.6013745704467355,
|
| 1645 |
+
"grad_norm": 0.6976660080607336,
|
| 1646 |
+
"learning_rate": 1.1313708031358183e-06,
|
| 1647 |
+
"loss": 0.0581,
|
| 1648 |
+
"step": 233
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.6082474226804124,
|
| 1652 |
+
"grad_norm": 0.5141954998445826,
|
| 1653 |
+
"learning_rate": 1.0935243153818437e-06,
|
| 1654 |
+
"loss": 0.0457,
|
| 1655 |
+
"step": 234
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.6151202749140894,
|
| 1659 |
+
"grad_norm": 0.6289297099798149,
|
| 1660 |
+
"learning_rate": 1.0562438044060846e-06,
|
| 1661 |
+
"loss": 0.0659,
|
| 1662 |
+
"step": 235
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.6219931271477663,
|
| 1666 |
+
"grad_norm": 0.6092162284865453,
|
| 1667 |
+
"learning_rate": 1.0195346714717813e-06,
|
| 1668 |
+
"loss": 0.0729,
|
| 1669 |
+
"step": 236
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.6288659793814433,
|
| 1673 |
+
"grad_norm": 0.6408567692863367,
|
| 1674 |
+
"learning_rate": 9.834022350599538e-07,
|
| 1675 |
+
"loss": 0.045,
|
| 1676 |
+
"step": 237
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.6357388316151202,
|
| 1680 |
+
"grad_norm": 0.7391843646157494,
|
| 1681 |
+
"learning_rate": 9.47851730098856e-07,
|
| 1682 |
+
"loss": 0.0909,
|
| 1683 |
+
"step": 238
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.6426116838487972,
|
| 1687 |
+
"grad_norm": 0.5707681560151466,
|
| 1688 |
+
"learning_rate": 9.128883072055411e-07,
|
| 1689 |
+
"loss": 0.0559,
|
| 1690 |
+
"step": 239
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.6494845360824741,
|
| 1694 |
+
"grad_norm": 0.5838006436632126,
|
| 1695 |
+
"learning_rate": 8.785170319396174e-07,
|
| 1696 |
+
"loss": 0.0605,
|
| 1697 |
+
"step": 240
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.656357388316151,
|
| 1701 |
+
"grad_norm": 0.6912410659973809,
|
| 1702 |
+
"learning_rate": 8.447428840693489e-07,
|
| 1703 |
+
"loss": 0.0682,
|
| 1704 |
+
"step": 241
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.663230240549828,
|
| 1708 |
+
"grad_norm": 0.4805724144435813,
|
| 1709 |
+
"learning_rate": 8.115707568501768e-07,
|
| 1710 |
+
"loss": 0.063,
|
| 1711 |
+
"step": 242
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.670103092783505,
|
| 1715 |
+
"grad_norm": 0.5101954556097166,
|
| 1716 |
+
"learning_rate": 7.790054563157745e-07,
|
| 1717 |
+
"loss": 0.0658,
|
| 1718 |
+
"step": 243
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.6769759450171822,
|
| 1722 |
+
"grad_norm": 0.6259841623930451,
|
| 1723 |
+
"learning_rate": 7.470517005817473e-07,
|
| 1724 |
+
"loss": 0.0659,
|
| 1725 |
+
"step": 244
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.6838487972508591,
|
| 1729 |
+
"grad_norm": 0.5974482851684669,
|
| 1730 |
+
"learning_rate": 7.157141191620548e-07,
|
| 1731 |
+
"loss": 0.0557,
|
| 1732 |
+
"step": 245
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.690721649484536,
|
| 1736 |
+
"grad_norm": 1.2520297398118274,
|
| 1737 |
+
"learning_rate": 6.849972522982845e-07,
|
| 1738 |
+
"loss": 0.0859,
|
| 1739 |
+
"step": 246
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.697594501718213,
|
| 1743 |
+
"grad_norm": 0.5413189690753024,
|
| 1744 |
+
"learning_rate": 6.549055503018575e-07,
|
| 1745 |
+
"loss": 0.0542,
|
| 1746 |
+
"step": 247
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.7044673539518902,
|
| 1750 |
+
"grad_norm": 0.5602612625442204,
|
| 1751 |
+
"learning_rate": 6.254433729092518e-07,
|
| 1752 |
+
"loss": 0.047,
|
| 1753 |
+
"step": 248
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.7113402061855671,
|
| 1757 |
+
"grad_norm": 0.5891723447509694,
|
| 1758 |
+
"learning_rate": 5.966149886503614e-07,
|
| 1759 |
+
"loss": 0.0543,
|
| 1760 |
+
"step": 249
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.718213058419244,
|
| 1764 |
+
"grad_norm": 0.524885223206303,
|
| 1765 |
+
"learning_rate": 5.684245742300625e-07,
|
| 1766 |
+
"loss": 0.0489,
|
| 1767 |
+
"step": 250
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.725085910652921,
|
| 1771 |
+
"grad_norm": 0.6230167881227805,
|
| 1772 |
+
"learning_rate": 5.408762139230889e-07,
|
| 1773 |
+
"loss": 0.068,
|
| 1774 |
+
"step": 251
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.731958762886598,
|
| 1778 |
+
"grad_norm": 0.6978111025044681,
|
| 1779 |
+
"learning_rate": 5.139738989822901e-07,
|
| 1780 |
+
"loss": 0.0669,
|
| 1781 |
+
"step": 252
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.738831615120275,
|
| 1785 |
+
"grad_norm": 0.581072545393766,
|
| 1786 |
+
"learning_rate": 4.877215270603752e-07,
|
| 1787 |
+
"loss": 0.0488,
|
| 1788 |
+
"step": 253
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 1.745704467353952,
|
| 1792 |
+
"grad_norm": 0.5243982772550911,
|
| 1793 |
+
"learning_rate": 4.6212290164521554e-07,
|
| 1794 |
+
"loss": 0.0385,
|
| 1795 |
+
"step": 254
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.7525773195876289,
|
| 1799 |
+
"grad_norm": 0.48908901752475303,
|
| 1800 |
+
"learning_rate": 4.371817315087845e-07,
|
| 1801 |
+
"loss": 0.047,
|
| 1802 |
+
"step": 255
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 1.7594501718213058,
|
| 1806 |
+
"grad_norm": 0.963772228189584,
|
| 1807 |
+
"learning_rate": 4.1290163016982855e-07,
|
| 1808 |
+
"loss": 0.044,
|
| 1809 |
+
"step": 256
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.7663230240549828,
|
| 1813 |
+
"grad_norm": 0.7161774828790537,
|
| 1814 |
+
"learning_rate": 3.8928611537033424e-07,
|
| 1815 |
+
"loss": 0.0617,
|
| 1816 |
+
"step": 257
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.7731958762886597,
|
| 1820 |
+
"grad_norm": 0.4782084753636899,
|
| 1821 |
+
"learning_rate": 3.663386085658693e-07,
|
| 1822 |
+
"loss": 0.042,
|
| 1823 |
+
"step": 258
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 1.7800687285223367,
|
| 1827 |
+
"grad_norm": 0.6211252920264383,
|
| 1828 |
+
"learning_rate": 3.4406243442987765e-07,
|
| 1829 |
+
"loss": 0.0677,
|
| 1830 |
+
"step": 259
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 1.7869415807560136,
|
| 1834 |
+
"grad_norm": 0.732301527572891,
|
| 1835 |
+
"learning_rate": 3.224608203719953e-07,
|
| 1836 |
+
"loss": 0.0657,
|
| 1837 |
+
"step": 260
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 1.7938144329896906,
|
| 1841 |
+
"grad_norm": 0.9715401095989267,
|
| 1842 |
+
"learning_rate": 3.015368960704584e-07,
|
| 1843 |
+
"loss": 0.0662,
|
| 1844 |
+
"step": 261
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 1.8006872852233677,
|
| 1848 |
+
"grad_norm": 0.6508304838866023,
|
| 1849 |
+
"learning_rate": 2.812936930186688e-07,
|
| 1850 |
+
"loss": 0.0477,
|
| 1851 |
+
"step": 262
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 1.8075601374570447,
|
| 1855 |
+
"grad_norm": 0.5348959997265367,
|
| 1856 |
+
"learning_rate": 2.617341440859883e-07,
|
| 1857 |
+
"loss": 0.0476,
|
| 1858 |
+
"step": 263
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 1.8144329896907216,
|
| 1862 |
+
"grad_norm": 0.4759300149556196,
|
| 1863 |
+
"learning_rate": 2.428610830928152e-07,
|
| 1864 |
+
"loss": 0.0518,
|
| 1865 |
+
"step": 264
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 1.8213058419243986,
|
| 1869 |
+
"grad_norm": 0.6144419146671524,
|
| 1870 |
+
"learning_rate": 2.2467724440002336e-07,
|
| 1871 |
+
"loss": 0.0566,
|
| 1872 |
+
"step": 265
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 1.8281786941580758,
|
| 1876 |
+
"grad_norm": 0.5190117212864495,
|
| 1877 |
+
"learning_rate": 2.0718526251279346e-07,
|
| 1878 |
+
"loss": 0.0521,
|
| 1879 |
+
"step": 266
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 1.8350515463917527,
|
| 1883 |
+
"grad_norm": 0.5319281331893329,
|
| 1884 |
+
"learning_rate": 1.9038767169893058e-07,
|
| 1885 |
+
"loss": 0.0492,
|
| 1886 |
+
"step": 267
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 1.8419243986254297,
|
| 1890 |
+
"grad_norm": 0.4999084380165005,
|
| 1891 |
+
"learning_rate": 1.7428690562169003e-07,
|
| 1892 |
+
"loss": 0.0632,
|
| 1893 |
+
"step": 268
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 1.8487972508591066,
|
| 1897 |
+
"grad_norm": 0.6515181928907401,
|
| 1898 |
+
"learning_rate": 1.5888529698718347e-07,
|
| 1899 |
+
"loss": 0.0555,
|
| 1900 |
+
"step": 269
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 1.8556701030927836,
|
| 1904 |
+
"grad_norm": 0.9433222847993904,
|
| 1905 |
+
"learning_rate": 1.4418507720641794e-07,
|
| 1906 |
+
"loss": 0.0558,
|
| 1907 |
+
"step": 270
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 1.8625429553264605,
|
| 1911 |
+
"grad_norm": 0.4954797355993427,
|
| 1912 |
+
"learning_rate": 1.3018837607199909e-07,
|
| 1913 |
+
"loss": 0.0512,
|
| 1914 |
+
"step": 271
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 1.8694158075601375,
|
| 1918 |
+
"grad_norm": 0.5329047083640734,
|
| 1919 |
+
"learning_rate": 1.1689722144956672e-07,
|
| 1920 |
+
"loss": 0.0512,
|
| 1921 |
+
"step": 272
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 1.8762886597938144,
|
| 1925 |
+
"grad_norm": 0.5224839088159605,
|
| 1926 |
+
"learning_rate": 1.0431353898399388e-07,
|
| 1927 |
+
"loss": 0.0546,
|
| 1928 |
+
"step": 273
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 1.8831615120274914,
|
| 1932 |
+
"grad_norm": 0.4909261130689138,
|
| 1933 |
+
"learning_rate": 9.243915182039431e-08,
|
| 1934 |
+
"loss": 0.0446,
|
| 1935 |
+
"step": 274
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 1.8900343642611683,
|
| 1939 |
+
"grad_norm": 0.4491028300117806,
|
| 1940 |
+
"learning_rate": 8.127578033998663e-08,
|
| 1941 |
+
"loss": 0.0441,
|
| 1942 |
+
"step": 275
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 1.8969072164948453,
|
| 1946 |
+
"grad_norm": 0.46103931455925695,
|
| 1947 |
+
"learning_rate": 7.082504191083417e-08,
|
| 1948 |
+
"loss": 0.0541,
|
| 1949 |
+
"step": 276
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 1.9037800687285222,
|
| 1953 |
+
"grad_norm": 0.6104853696569538,
|
| 1954 |
+
"learning_rate": 6.108845065352864e-08,
|
| 1955 |
+
"loss": 0.0586,
|
| 1956 |
+
"step": 277
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 1.9106529209621992,
|
| 1960 |
+
"grad_norm": 0.6027793230536882,
|
| 1961 |
+
"learning_rate": 5.206741722181385e-08,
|
| 1962 |
+
"loss": 0.056,
|
| 1963 |
+
"step": 278
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 1.9175257731958761,
|
| 1967 |
+
"grad_norm": 0.5563596839626086,
|
| 1968 |
+
"learning_rate": 4.376324859820924e-08,
|
| 1969 |
+
"loss": 0.0562,
|
| 1970 |
+
"step": 279
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 1.9243986254295533,
|
| 1974 |
+
"grad_norm": 0.5656167288368498,
|
| 1975 |
+
"learning_rate": 3.617714790465576e-08,
|
| 1976 |
+
"loss": 0.0533,
|
| 1977 |
+
"step": 280
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 1.9312714776632303,
|
| 1981 |
+
"grad_norm": 0.4650464709245068,
|
| 1982 |
+
"learning_rate": 2.9310214228202016e-08,
|
| 1983 |
+
"loss": 0.0474,
|
| 1984 |
+
"step": 281
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 1.9381443298969072,
|
| 1988 |
+
"grad_norm": 0.48539289078871045,
|
| 1989 |
+
"learning_rate": 2.3163442461766604e-08,
|
| 1990 |
+
"loss": 0.0453,
|
| 1991 |
+
"step": 282
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 1.9450171821305842,
|
| 1995 |
+
"grad_norm": 0.6055827187788138,
|
| 1996 |
+
"learning_rate": 1.7737723159999e-08,
|
| 1997 |
+
"loss": 0.063,
|
| 1998 |
+
"step": 283
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 1.9518900343642611,
|
| 2002 |
+
"grad_norm": 0.4522026290029071,
|
| 2003 |
+
"learning_rate": 1.3033842410251074e-08,
|
| 2004 |
+
"loss": 0.0444,
|
| 2005 |
+
"step": 284
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 1.9587628865979383,
|
| 2009 |
+
"grad_norm": 0.7318298157406874,
|
| 2010 |
+
"learning_rate": 9.052481718690998e-09,
|
| 2011 |
+
"loss": 0.0519,
|
| 2012 |
+
"step": 285
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 1.9656357388316152,
|
| 2016 |
+
"grad_norm": 0.5726700268602862,
|
| 2017 |
+
"learning_rate": 5.794217911562205e-09,
|
| 2018 |
+
"loss": 0.0519,
|
| 2019 |
+
"step": 286
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 1.9725085910652922,
|
| 2023 |
+
"grad_norm": 0.46308098209585613,
|
| 2024 |
+
"learning_rate": 3.2595230516152543e-09,
|
| 2025 |
+
"loss": 0.0443,
|
| 2026 |
+
"step": 287
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 1.9793814432989691,
|
| 2030 |
+
"grad_norm": 0.6080597442162697,
|
| 2031 |
+
"learning_rate": 1.4487643697103092e-09,
|
| 2032 |
+
"loss": 0.0656,
|
| 2033 |
+
"step": 288
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 1.986254295532646,
|
| 2037 |
+
"grad_norm": 0.4431036720666356,
|
| 2038 |
+
"learning_rate": 3.6220421161692333e-10,
|
| 2039 |
+
"loss": 0.0363,
|
| 2040 |
+
"step": 289
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 1.993127147766323,
|
| 2044 |
+
"grad_norm": 0.5380313687107504,
|
| 2045 |
+
"learning_rate": 0.0,
|
| 2046 |
+
"loss": 0.0658,
|
| 2047 |
+
"step": 290
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 1.993127147766323,
|
| 2051 |
+
"eval_loss": 0.06153901666402817,
|
| 2052 |
+
"eval_runtime": 28.2674,
|
| 2053 |
+
"eval_samples_per_second": 34.634,
|
| 2054 |
+
"eval_steps_per_second": 1.097,
|
| 2055 |
+
"step": 290
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 1.993127147766323,
|
| 2059 |
+
"step": 290,
|
| 2060 |
+
"total_flos": 8.781509506919629e+16,
|
| 2061 |
+
"train_loss": 0.083118896693762,
|
| 2062 |
+
"train_runtime": 3358.7055,
|
| 2063 |
+
"train_samples_per_second": 11.075,
|
| 2064 |
+
"train_steps_per_second": 0.086
|
| 2065 |
+
}
|
| 2066 |
+
],
|
| 2067 |
+
"logging_steps": 1,
|
| 2068 |
+
"max_steps": 290,
|
| 2069 |
+
"num_input_tokens_seen": 0,
|
| 2070 |
+
"num_train_epochs": 2,
|
| 2071 |
+
"save_steps": 500,
|
| 2072 |
+
"stateful_callbacks": {
|
| 2073 |
+
"TrainerControl": {
|
| 2074 |
+
"args": {
|
| 2075 |
+
"should_epoch_stop": false,
|
| 2076 |
+
"should_evaluate": false,
|
| 2077 |
+
"should_log": false,
|
| 2078 |
+
"should_save": true,
|
| 2079 |
+
"should_training_stop": true
|
| 2080 |
+
},
|
| 2081 |
+
"attributes": {}
|
| 2082 |
+
}
|
| 2083 |
+
},
|
| 2084 |
+
"total_flos": 8.781509506919629e+16,
|
| 2085 |
+
"train_batch_size": 8,
|
| 2086 |
+
"trial_name": null,
|
| 2087 |
+
"trial_params": null
|
| 2088 |
+
}
|