Commit
•
331636a
1
Parent(s):
fb326cf
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 +2158 -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_seed25
|
| 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_seed25
|
| 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: 0.0608
|
| 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 |
+
| 0.0683 | 0.9967 | 150 | 0.0725 |
|
| 59 |
+
| 0.0482 | 1.9934 | 300 | 0.0608 |
|
| 60 |
+
|
| 61 |
+
|
| 62 |
+
### Framework versions
|
| 63 |
+
|
| 64 |
+
- Transformers 4.45.0.dev0
|
| 65 |
+
- Pytorch 2.4.0+cu121
|
| 66 |
+
- Datasets 3.0.1
|
| 67 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.9933554817275747,
|
| 3 |
+
"total_flos": 1.1631033437569024e+17,
|
| 4 |
+
"train_loss": 0.07701069086790086,
|
| 5 |
+
"train_runtime": 4589.0959,
|
| 6 |
+
"train_samples": 19216,
|
| 7 |
+
"train_samples_per_second": 8.375,
|
| 8 |
+
"train_steps_per_second": 0.065
|
| 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": 1.9933554817275747,
|
| 3 |
+
"total_flos": 1.1631033437569024e+17,
|
| 4 |
+
"train_loss": 0.07701069086790086,
|
| 5 |
+
"train_runtime": 4589.0959,
|
| 6 |
+
"train_samples": 19216,
|
| 7 |
+
"train_samples_per_second": 8.375,
|
| 8 |
+
"train_steps_per_second": 0.065
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2158 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.9933554817275747,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 300,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.006644518272425249,
|
| 13 |
+
"grad_norm": 10.145524315941286,
|
| 14 |
+
"learning_rate": 3.3333333333333335e-07,
|
| 15 |
+
"loss": 0.2861,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.013289036544850499,
|
| 20 |
+
"grad_norm": 8.915211114203709,
|
| 21 |
+
"learning_rate": 6.666666666666667e-07,
|
| 22 |
+
"loss": 0.2616,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.019933554817275746,
|
| 27 |
+
"grad_norm": 8.777900140983958,
|
| 28 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 29 |
+
"loss": 0.2835,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.026578073089700997,
|
| 34 |
+
"grad_norm": 8.285488875547399,
|
| 35 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 36 |
+
"loss": 0.2231,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.03322259136212625,
|
| 41 |
+
"grad_norm": 7.810022519562283,
|
| 42 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 43 |
+
"loss": 0.2467,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.03986710963455149,
|
| 48 |
+
"grad_norm": 6.560410496844325,
|
| 49 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 50 |
+
"loss": 0.241,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.046511627906976744,
|
| 55 |
+
"grad_norm": 4.7170552839011775,
|
| 56 |
+
"learning_rate": 2.3333333333333336e-06,
|
| 57 |
+
"loss": 0.2127,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.053156146179401995,
|
| 62 |
+
"grad_norm": 4.183129818287662,
|
| 63 |
+
"learning_rate": 2.666666666666667e-06,
|
| 64 |
+
"loss": 0.2308,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.059800664451827246,
|
| 69 |
+
"grad_norm": 3.1610581594177942,
|
| 70 |
+
"learning_rate": 3e-06,
|
| 71 |
+
"loss": 0.1254,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0664451827242525,
|
| 76 |
+
"grad_norm": 2.538388610366679,
|
| 77 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 78 |
+
"loss": 0.178,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.07308970099667775,
|
| 83 |
+
"grad_norm": 1.6417414573531897,
|
| 84 |
+
"learning_rate": 3.6666666666666666e-06,
|
| 85 |
+
"loss": 0.1305,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.07973421926910298,
|
| 90 |
+
"grad_norm": 5.9298971216598115,
|
| 91 |
+
"learning_rate": 4.000000000000001e-06,
|
| 92 |
+
"loss": 0.152,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.08637873754152824,
|
| 97 |
+
"grad_norm": 3.0446940552354858,
|
| 98 |
+
"learning_rate": 4.333333333333334e-06,
|
| 99 |
+
"loss": 0.1014,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.09302325581395349,
|
| 104 |
+
"grad_norm": 2.2106089020081106,
|
| 105 |
+
"learning_rate": 4.666666666666667e-06,
|
| 106 |
+
"loss": 0.1222,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.09966777408637874,
|
| 111 |
+
"grad_norm": 2.547231850213915,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.1536,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.10631229235880399,
|
| 118 |
+
"grad_norm": 1.9791098486011387,
|
| 119 |
+
"learning_rate": 5.333333333333334e-06,
|
| 120 |
+
"loss": 0.1499,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.11295681063122924,
|
| 125 |
+
"grad_norm": 1.5927564141097508,
|
| 126 |
+
"learning_rate": 5.666666666666667e-06,
|
| 127 |
+
"loss": 0.0975,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.11960132890365449,
|
| 132 |
+
"grad_norm": 1.242283265980379,
|
| 133 |
+
"learning_rate": 6e-06,
|
| 134 |
+
"loss": 0.0974,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.12624584717607973,
|
| 139 |
+
"grad_norm": 2.0819749599134756,
|
| 140 |
+
"learning_rate": 6.333333333333333e-06,
|
| 141 |
+
"loss": 0.105,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.132890365448505,
|
| 146 |
+
"grad_norm": 3.1481612101400835,
|
| 147 |
+
"learning_rate": 6.666666666666667e-06,
|
| 148 |
+
"loss": 0.1413,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.13953488372093023,
|
| 153 |
+
"grad_norm": 1.329108390047055,
|
| 154 |
+
"learning_rate": 7e-06,
|
| 155 |
+
"loss": 0.082,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.1461794019933555,
|
| 160 |
+
"grad_norm": 2.205626574560857,
|
| 161 |
+
"learning_rate": 7.333333333333333e-06,
|
| 162 |
+
"loss": 0.098,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.15282392026578073,
|
| 167 |
+
"grad_norm": 1.323315317507035,
|
| 168 |
+
"learning_rate": 7.666666666666667e-06,
|
| 169 |
+
"loss": 0.0828,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.15946843853820597,
|
| 174 |
+
"grad_norm": 1.47807340385477,
|
| 175 |
+
"learning_rate": 8.000000000000001e-06,
|
| 176 |
+
"loss": 0.1228,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.16611295681063123,
|
| 181 |
+
"grad_norm": 1.4244378567157914,
|
| 182 |
+
"learning_rate": 8.333333333333334e-06,
|
| 183 |
+
"loss": 0.093,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.17275747508305647,
|
| 188 |
+
"grad_norm": 1.6965216293403562,
|
| 189 |
+
"learning_rate": 8.666666666666668e-06,
|
| 190 |
+
"loss": 0.1138,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.17940199335548174,
|
| 195 |
+
"grad_norm": 2.4560093523672966,
|
| 196 |
+
"learning_rate": 9e-06,
|
| 197 |
+
"loss": 0.1197,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.18604651162790697,
|
| 202 |
+
"grad_norm": 1.920641163538201,
|
| 203 |
+
"learning_rate": 9.333333333333334e-06,
|
| 204 |
+
"loss": 0.1118,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.19269102990033224,
|
| 209 |
+
"grad_norm": 2.4139931857560577,
|
| 210 |
+
"learning_rate": 9.666666666666667e-06,
|
| 211 |
+
"loss": 0.1143,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.19933554817275748,
|
| 216 |
+
"grad_norm": 1.212761719437112,
|
| 217 |
+
"learning_rate": 1e-05,
|
| 218 |
+
"loss": 0.1061,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.2059800664451827,
|
| 223 |
+
"grad_norm": 2.647502675755485,
|
| 224 |
+
"learning_rate": 9.999661540018812e-06,
|
| 225 |
+
"loss": 0.115,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.21262458471760798,
|
| 230 |
+
"grad_norm": 1.446671287082578,
|
| 231 |
+
"learning_rate": 9.99864620589731e-06,
|
| 232 |
+
"loss": 0.1063,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.21926910299003322,
|
| 237 |
+
"grad_norm": 1.1989767107611067,
|
| 238 |
+
"learning_rate": 9.99695413509548e-06,
|
| 239 |
+
"loss": 0.0783,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.22591362126245848,
|
| 244 |
+
"grad_norm": 1.5133023884865615,
|
| 245 |
+
"learning_rate": 9.994585556692624e-06,
|
| 246 |
+
"loss": 0.0914,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.23255813953488372,
|
| 251 |
+
"grad_norm": 1.4282681068359147,
|
| 252 |
+
"learning_rate": 9.991540791356342e-06,
|
| 253 |
+
"loss": 0.1023,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.23920265780730898,
|
| 258 |
+
"grad_norm": 1.5675540747452452,
|
| 259 |
+
"learning_rate": 9.987820251299121e-06,
|
| 260 |
+
"loss": 0.0737,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.24584717607973422,
|
| 265 |
+
"grad_norm": 1.4699450317353462,
|
| 266 |
+
"learning_rate": 9.98342444022253e-06,
|
| 267 |
+
"loss": 0.1156,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.25249169435215946,
|
| 272 |
+
"grad_norm": 2.2211547112198873,
|
| 273 |
+
"learning_rate": 9.978353953249023e-06,
|
| 274 |
+
"loss": 0.1114,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.2591362126245847,
|
| 279 |
+
"grad_norm": 1.1602597727805404,
|
| 280 |
+
"learning_rate": 9.972609476841368e-06,
|
| 281 |
+
"loss": 0.1032,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.26578073089701,
|
| 286 |
+
"grad_norm": 0.9637077068719486,
|
| 287 |
+
"learning_rate": 9.966191788709716e-06,
|
| 288 |
+
"loss": 0.0681,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.2724252491694352,
|
| 293 |
+
"grad_norm": 1.6293839524249614,
|
| 294 |
+
"learning_rate": 9.959101757706308e-06,
|
| 295 |
+
"loss": 0.115,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.27906976744186046,
|
| 300 |
+
"grad_norm": 0.9732726248145027,
|
| 301 |
+
"learning_rate": 9.951340343707852e-06,
|
| 302 |
+
"loss": 0.0975,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.2857142857142857,
|
| 307 |
+
"grad_norm": 1.9370817624388006,
|
| 308 |
+
"learning_rate": 9.942908597485558e-06,
|
| 309 |
+
"loss": 0.1148,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.292358803986711,
|
| 314 |
+
"grad_norm": 1.6614214400661733,
|
| 315 |
+
"learning_rate": 9.933807660562898e-06,
|
| 316 |
+
"loss": 0.1163,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.29900332225913623,
|
| 321 |
+
"grad_norm": 1.1978664226300189,
|
| 322 |
+
"learning_rate": 9.924038765061042e-06,
|
| 323 |
+
"loss": 0.0984,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.30564784053156147,
|
| 328 |
+
"grad_norm": 1.3623012344086183,
|
| 329 |
+
"learning_rate": 9.913603233532067e-06,
|
| 330 |
+
"loss": 0.0993,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.3122923588039867,
|
| 335 |
+
"grad_norm": 0.987057722774455,
|
| 336 |
+
"learning_rate": 9.902502478779897e-06,
|
| 337 |
+
"loss": 0.0554,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.31893687707641194,
|
| 342 |
+
"grad_norm": 1.748369907296719,
|
| 343 |
+
"learning_rate": 9.890738003669029e-06,
|
| 344 |
+
"loss": 0.1239,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.32558139534883723,
|
| 349 |
+
"grad_norm": 2.491434672974267,
|
| 350 |
+
"learning_rate": 9.878311400921072e-06,
|
| 351 |
+
"loss": 0.1029,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.33222591362126247,
|
| 356 |
+
"grad_norm": 1.4006160903691325,
|
| 357 |
+
"learning_rate": 9.86522435289912e-06,
|
| 358 |
+
"loss": 0.0996,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.3388704318936877,
|
| 363 |
+
"grad_norm": 2.2756678251100726,
|
| 364 |
+
"learning_rate": 9.851478631379982e-06,
|
| 365 |
+
"loss": 0.1372,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.34551495016611294,
|
| 370 |
+
"grad_norm": 1.3072647570291376,
|
| 371 |
+
"learning_rate": 9.83707609731432e-06,
|
| 372 |
+
"loss": 0.0867,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.3521594684385382,
|
| 377 |
+
"grad_norm": 1.8929458494674647,
|
| 378 |
+
"learning_rate": 9.822018700574696e-06,
|
| 379 |
+
"loss": 0.131,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.3588039867109635,
|
| 384 |
+
"grad_norm": 1.3900822961021344,
|
| 385 |
+
"learning_rate": 9.806308479691595e-06,
|
| 386 |
+
"loss": 0.0702,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.3654485049833887,
|
| 391 |
+
"grad_norm": 0.7898540133946153,
|
| 392 |
+
"learning_rate": 9.789947561577445e-06,
|
| 393 |
+
"loss": 0.0857,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.37209302325581395,
|
| 398 |
+
"grad_norm": 1.8444881635021309,
|
| 399 |
+
"learning_rate": 9.77293816123866e-06,
|
| 400 |
+
"loss": 0.1006,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.3787375415282392,
|
| 405 |
+
"grad_norm": 1.1604374728958515,
|
| 406 |
+
"learning_rate": 9.755282581475769e-06,
|
| 407 |
+
"loss": 0.077,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.3853820598006645,
|
| 412 |
+
"grad_norm": 1.361789017794424,
|
| 413 |
+
"learning_rate": 9.736983212571646e-06,
|
| 414 |
+
"loss": 0.0755,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.3920265780730897,
|
| 419 |
+
"grad_norm": 1.6989317046427177,
|
| 420 |
+
"learning_rate": 9.718042531967918e-06,
|
| 421 |
+
"loss": 0.1073,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.39867109634551495,
|
| 426 |
+
"grad_norm": 0.5576923228958391,
|
| 427 |
+
"learning_rate": 9.698463103929542e-06,
|
| 428 |
+
"loss": 0.0563,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.4053156146179402,
|
| 433 |
+
"grad_norm": 1.197144883948249,
|
| 434 |
+
"learning_rate": 9.678247579197658e-06,
|
| 435 |
+
"loss": 0.0835,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.4119601328903654,
|
| 440 |
+
"grad_norm": 0.7800865622678685,
|
| 441 |
+
"learning_rate": 9.657398694630713e-06,
|
| 442 |
+
"loss": 0.084,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.4186046511627907,
|
| 447 |
+
"grad_norm": 1.332148923735981,
|
| 448 |
+
"learning_rate": 9.635919272833938e-06,
|
| 449 |
+
"loss": 0.0965,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.42524916943521596,
|
| 454 |
+
"grad_norm": 0.9537850616235446,
|
| 455 |
+
"learning_rate": 9.613812221777212e-06,
|
| 456 |
+
"loss": 0.0784,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.4318936877076412,
|
| 461 |
+
"grad_norm": 0.9773777024805813,
|
| 462 |
+
"learning_rate": 9.591080534401371e-06,
|
| 463 |
+
"loss": 0.094,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.43853820598006643,
|
| 468 |
+
"grad_norm": 0.9622850200719331,
|
| 469 |
+
"learning_rate": 9.567727288213005e-06,
|
| 470 |
+
"loss": 0.0682,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.44518272425249167,
|
| 475 |
+
"grad_norm": 1.1749521195263826,
|
| 476 |
+
"learning_rate": 9.543755644867823e-06,
|
| 477 |
+
"loss": 0.1174,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.45182724252491696,
|
| 482 |
+
"grad_norm": 0.9900437964371273,
|
| 483 |
+
"learning_rate": 9.519168849742603e-06,
|
| 484 |
+
"loss": 0.0631,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.4584717607973422,
|
| 489 |
+
"grad_norm": 1.1369496264217245,
|
| 490 |
+
"learning_rate": 9.493970231495836e-06,
|
| 491 |
+
"loss": 0.091,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.46511627906976744,
|
| 496 |
+
"grad_norm": 0.7596879303837656,
|
| 497 |
+
"learning_rate": 9.468163201617063e-06,
|
| 498 |
+
"loss": 0.0737,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.4717607973421927,
|
| 503 |
+
"grad_norm": 2.091451100542592,
|
| 504 |
+
"learning_rate": 9.441751253965022e-06,
|
| 505 |
+
"loss": 0.1005,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.47840531561461797,
|
| 510 |
+
"grad_norm": 1.093842023272068,
|
| 511 |
+
"learning_rate": 9.414737964294636e-06,
|
| 512 |
+
"loss": 0.0777,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.4850498338870432,
|
| 517 |
+
"grad_norm": 0.9952555604898301,
|
| 518 |
+
"learning_rate": 9.38712698977291e-06,
|
| 519 |
+
"loss": 0.0877,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.49169435215946844,
|
| 524 |
+
"grad_norm": 1.1620904184349503,
|
| 525 |
+
"learning_rate": 9.358922068483813e-06,
|
| 526 |
+
"loss": 0.0944,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.4983388704318937,
|
| 531 |
+
"grad_norm": 0.5355560711239664,
|
| 532 |
+
"learning_rate": 9.330127018922195e-06,
|
| 533 |
+
"loss": 0.0655,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.5049833887043189,
|
| 538 |
+
"grad_norm": 1.0977107361921403,
|
| 539 |
+
"learning_rate": 9.30074573947683e-06,
|
| 540 |
+
"loss": 0.0837,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.5116279069767442,
|
| 545 |
+
"grad_norm": 0.5247471981497749,
|
| 546 |
+
"learning_rate": 9.27078220790263e-06,
|
| 547 |
+
"loss": 0.0593,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.5182724252491694,
|
| 552 |
+
"grad_norm": 0.5701004570093748,
|
| 553 |
+
"learning_rate": 9.24024048078213e-06,
|
| 554 |
+
"loss": 0.0459,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.5249169435215947,
|
| 559 |
+
"grad_norm": 1.0366232450474129,
|
| 560 |
+
"learning_rate": 9.209124692976287e-06,
|
| 561 |
+
"loss": 0.0876,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.53156146179402,
|
| 566 |
+
"grad_norm": 0.6364682974892603,
|
| 567 |
+
"learning_rate": 9.177439057064684e-06,
|
| 568 |
+
"loss": 0.0763,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.5382059800664452,
|
| 573 |
+
"grad_norm": 1.0568589661747034,
|
| 574 |
+
"learning_rate": 9.145187862775208e-06,
|
| 575 |
+
"loss": 0.093,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.5448504983388704,
|
| 580 |
+
"grad_norm": 0.9386306410175308,
|
| 581 |
+
"learning_rate": 9.112375476403313e-06,
|
| 582 |
+
"loss": 0.0891,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.5514950166112956,
|
| 587 |
+
"grad_norm": 0.8643230084646484,
|
| 588 |
+
"learning_rate": 9.079006340220862e-06,
|
| 589 |
+
"loss": 0.096,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.5581395348837209,
|
| 594 |
+
"grad_norm": 0.6462165981943299,
|
| 595 |
+
"learning_rate": 9.045084971874738e-06,
|
| 596 |
+
"loss": 0.0595,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.5647840531561462,
|
| 601 |
+
"grad_norm": 0.7897404631218589,
|
| 602 |
+
"learning_rate": 9.01061596377522e-06,
|
| 603 |
+
"loss": 0.0731,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.5714285714285714,
|
| 608 |
+
"grad_norm": 1.0476593444440303,
|
| 609 |
+
"learning_rate": 8.97560398247424e-06,
|
| 610 |
+
"loss": 0.0705,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.5780730897009967,
|
| 615 |
+
"grad_norm": 0.7346838336829987,
|
| 616 |
+
"learning_rate": 8.94005376803361e-06,
|
| 617 |
+
"loss": 0.0749,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.584717607973422,
|
| 622 |
+
"grad_norm": 1.1640993902217747,
|
| 623 |
+
"learning_rate": 8.903970133383297e-06,
|
| 624 |
+
"loss": 0.0993,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.5913621262458472,
|
| 629 |
+
"grad_norm": 0.8668509799741243,
|
| 630 |
+
"learning_rate": 8.867357963669821e-06,
|
| 631 |
+
"loss": 0.0781,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.5980066445182725,
|
| 636 |
+
"grad_norm": 0.8019269925738027,
|
| 637 |
+
"learning_rate": 8.83022221559489e-06,
|
| 638 |
+
"loss": 0.0627,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.6046511627906976,
|
| 643 |
+
"grad_norm": 0.9309523320951085,
|
| 644 |
+
"learning_rate": 8.792567916744346e-06,
|
| 645 |
+
"loss": 0.0689,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.6112956810631229,
|
| 650 |
+
"grad_norm": 0.9960236498367079,
|
| 651 |
+
"learning_rate": 8.754400164907496e-06,
|
| 652 |
+
"loss": 0.0817,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.6179401993355482,
|
| 657 |
+
"grad_norm": 1.0124963384933403,
|
| 658 |
+
"learning_rate": 8.715724127386971e-06,
|
| 659 |
+
"loss": 0.0917,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.6245847176079734,
|
| 664 |
+
"grad_norm": 0.9034419295571391,
|
| 665 |
+
"learning_rate": 8.676545040299145e-06,
|
| 666 |
+
"loss": 0.0913,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.6312292358803987,
|
| 671 |
+
"grad_norm": 0.7241319583554866,
|
| 672 |
+
"learning_rate": 8.636868207865244e-06,
|
| 673 |
+
"loss": 0.0758,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.6378737541528239,
|
| 678 |
+
"grad_norm": 0.9291976143638696,
|
| 679 |
+
"learning_rate": 8.596699001693257e-06,
|
| 680 |
+
"loss": 0.0993,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.6445182724252492,
|
| 685 |
+
"grad_norm": 0.7063222766334012,
|
| 686 |
+
"learning_rate": 8.556042860050686e-06,
|
| 687 |
+
"loss": 0.087,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.6511627906976745,
|
| 692 |
+
"grad_norm": 0.7828811245723147,
|
| 693 |
+
"learning_rate": 8.51490528712831e-06,
|
| 694 |
+
"loss": 0.0748,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.6578073089700996,
|
| 699 |
+
"grad_norm": 0.7179259078532849,
|
| 700 |
+
"learning_rate": 8.473291852294986e-06,
|
| 701 |
+
"loss": 0.0708,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.6644518272425249,
|
| 706 |
+
"grad_norm": 1.0029460084879072,
|
| 707 |
+
"learning_rate": 8.43120818934367e-06,
|
| 708 |
+
"loss": 0.0851,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.6710963455149501,
|
| 713 |
+
"grad_norm": 0.8100677095349954,
|
| 714 |
+
"learning_rate": 8.388659995728662e-06,
|
| 715 |
+
"loss": 0.0965,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.6777408637873754,
|
| 720 |
+
"grad_norm": 0.7387893774671513,
|
| 721 |
+
"learning_rate": 8.345653031794292e-06,
|
| 722 |
+
"loss": 0.0887,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.6843853820598007,
|
| 727 |
+
"grad_norm": 0.8786303341962616,
|
| 728 |
+
"learning_rate": 8.302193119995038e-06,
|
| 729 |
+
"loss": 0.0833,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.6910299003322259,
|
| 734 |
+
"grad_norm": 0.6498104624184341,
|
| 735 |
+
"learning_rate": 8.258286144107277e-06,
|
| 736 |
+
"loss": 0.0664,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.6976744186046512,
|
| 741 |
+
"grad_norm": 0.8053829136759763,
|
| 742 |
+
"learning_rate": 8.213938048432697e-06,
|
| 743 |
+
"loss": 0.0773,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.7043189368770764,
|
| 748 |
+
"grad_norm": 0.7440146521214637,
|
| 749 |
+
"learning_rate": 8.16915483699355e-06,
|
| 750 |
+
"loss": 0.084,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.7109634551495017,
|
| 755 |
+
"grad_norm": 0.8430477204164517,
|
| 756 |
+
"learning_rate": 8.123942572719801e-06,
|
| 757 |
+
"loss": 0.0754,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.717607973421927,
|
| 762 |
+
"grad_norm": 0.7844336038759876,
|
| 763 |
+
"learning_rate": 8.078307376628292e-06,
|
| 764 |
+
"loss": 0.0681,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.7242524916943521,
|
| 769 |
+
"grad_norm": 0.6865608703175325,
|
| 770 |
+
"learning_rate": 8.032255426994069e-06,
|
| 771 |
+
"loss": 0.0808,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.7308970099667774,
|
| 776 |
+
"grad_norm": 0.8626677757680601,
|
| 777 |
+
"learning_rate": 7.985792958513932e-06,
|
| 778 |
+
"loss": 0.1014,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.7375415282392026,
|
| 783 |
+
"grad_norm": 0.6458902559836811,
|
| 784 |
+
"learning_rate": 7.938926261462366e-06,
|
| 785 |
+
"loss": 0.0727,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.7441860465116279,
|
| 790 |
+
"grad_norm": 0.67871461818545,
|
| 791 |
+
"learning_rate": 7.891661680839932e-06,
|
| 792 |
+
"loss": 0.0658,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.7508305647840532,
|
| 797 |
+
"grad_norm": 0.7537117403397503,
|
| 798 |
+
"learning_rate": 7.84400561551426e-06,
|
| 799 |
+
"loss": 0.0826,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.7574750830564784,
|
| 804 |
+
"grad_norm": 0.849811513648714,
|
| 805 |
+
"learning_rate": 7.795964517353734e-06,
|
| 806 |
+
"loss": 0.0795,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.7641196013289037,
|
| 811 |
+
"grad_norm": 0.6434504848546716,
|
| 812 |
+
"learning_rate": 7.747544890354031e-06,
|
| 813 |
+
"loss": 0.0661,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.770764119601329,
|
| 818 |
+
"grad_norm": 0.8079608133403346,
|
| 819 |
+
"learning_rate": 7.698753289757565e-06,
|
| 820 |
+
"loss": 0.075,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.7774086378737541,
|
| 825 |
+
"grad_norm": 0.9390957260974234,
|
| 826 |
+
"learning_rate": 7.649596321166024e-06,
|
| 827 |
+
"loss": 0.0649,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.7840531561461794,
|
| 832 |
+
"grad_norm": 0.8258259729749267,
|
| 833 |
+
"learning_rate": 7.600080639646077e-06,
|
| 834 |
+
"loss": 0.076,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.7906976744186046,
|
| 839 |
+
"grad_norm": 0.8344165033408021,
|
| 840 |
+
"learning_rate": 7.550212948828377e-06,
|
| 841 |
+
"loss": 0.0658,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.7973421926910299,
|
| 846 |
+
"grad_norm": 0.7468613999721869,
|
| 847 |
+
"learning_rate": 7.500000000000001e-06,
|
| 848 |
+
"loss": 0.0731,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.8039867109634552,
|
| 853 |
+
"grad_norm": 0.7375954465469966,
|
| 854 |
+
"learning_rate": 7.449448591190436e-06,
|
| 855 |
+
"loss": 0.0775,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.8106312292358804,
|
| 860 |
+
"grad_norm": 0.5770159108524059,
|
| 861 |
+
"learning_rate": 7.398565566251232e-06,
|
| 862 |
+
"loss": 0.0766,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.8172757475083057,
|
| 867 |
+
"grad_norm": 0.6213954523156613,
|
| 868 |
+
"learning_rate": 7.347357813929455e-06,
|
| 869 |
+
"loss": 0.0684,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.8239202657807309,
|
| 874 |
+
"grad_norm": 0.6319841085682676,
|
| 875 |
+
"learning_rate": 7.295832266935059e-06,
|
| 876 |
+
"loss": 0.0508,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.8305647840531561,
|
| 881 |
+
"grad_norm": 0.5561773891072708,
|
| 882 |
+
"learning_rate": 7.243995901002312e-06,
|
| 883 |
+
"loss": 0.0416,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.8372093023255814,
|
| 888 |
+
"grad_norm": 0.6768768843907513,
|
| 889 |
+
"learning_rate": 7.191855733945388e-06,
|
| 890 |
+
"loss": 0.1001,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.8438538205980066,
|
| 895 |
+
"grad_norm": 0.8870781137917316,
|
| 896 |
+
"learning_rate": 7.1394188247082715e-06,
|
| 897 |
+
"loss": 0.096,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.8504983388704319,
|
| 902 |
+
"grad_norm": 0.6199813951371215,
|
| 903 |
+
"learning_rate": 7.08669227240909e-06,
|
| 904 |
+
"loss": 0.058,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.8571428571428571,
|
| 909 |
+
"grad_norm": 0.6258631411773975,
|
| 910 |
+
"learning_rate": 7.033683215379002e-06,
|
| 911 |
+
"loss": 0.0797,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.8637873754152824,
|
| 916 |
+
"grad_norm": 0.7141241764483236,
|
| 917 |
+
"learning_rate": 6.980398830195785e-06,
|
| 918 |
+
"loss": 0.0616,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.8704318936877077,
|
| 923 |
+
"grad_norm": 0.6539814755788471,
|
| 924 |
+
"learning_rate": 6.9268463307122425e-06,
|
| 925 |
+
"loss": 0.0543,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.8770764119601329,
|
| 930 |
+
"grad_norm": 0.6696475606516166,
|
| 931 |
+
"learning_rate": 6.873032967079562e-06,
|
| 932 |
+
"loss": 0.0922,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.8837209302325582,
|
| 937 |
+
"grad_norm": 0.7256752618037919,
|
| 938 |
+
"learning_rate": 6.818966024765758e-06,
|
| 939 |
+
"loss": 0.0723,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.8903654485049833,
|
| 944 |
+
"grad_norm": 0.7949722457642716,
|
| 945 |
+
"learning_rate": 6.7646528235693445e-06,
|
| 946 |
+
"loss": 0.0919,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.8970099667774086,
|
| 951 |
+
"grad_norm": 0.595409116890567,
|
| 952 |
+
"learning_rate": 6.710100716628345e-06,
|
| 953 |
+
"loss": 0.0585,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.9036544850498339,
|
| 958 |
+
"grad_norm": 1.0033697089937976,
|
| 959 |
+
"learning_rate": 6.655317089424791e-06,
|
| 960 |
+
"loss": 0.1081,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.9102990033222591,
|
| 965 |
+
"grad_norm": 0.710716471651743,
|
| 966 |
+
"learning_rate": 6.600309358784858e-06,
|
| 967 |
+
"loss": 0.0761,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.9169435215946844,
|
| 972 |
+
"grad_norm": 0.576448742099724,
|
| 973 |
+
"learning_rate": 6.545084971874738e-06,
|
| 974 |
+
"loss": 0.0614,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.9235880398671097,
|
| 979 |
+
"grad_norm": 0.8348391733283355,
|
| 980 |
+
"learning_rate": 6.48965140519241e-06,
|
| 981 |
+
"loss": 0.063,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.9302325581395349,
|
| 986 |
+
"grad_norm": 1.1013542545275912,
|
| 987 |
+
"learning_rate": 6.434016163555452e-06,
|
| 988 |
+
"loss": 0.0734,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.9368770764119602,
|
| 993 |
+
"grad_norm": 0.779239872251012,
|
| 994 |
+
"learning_rate": 6.378186779084996e-06,
|
| 995 |
+
"loss": 0.0721,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.9435215946843853,
|
| 1000 |
+
"grad_norm": 0.6235545197422645,
|
| 1001 |
+
"learning_rate": 6.322170810186013e-06,
|
| 1002 |
+
"loss": 0.0623,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.9501661129568106,
|
| 1007 |
+
"grad_norm": 0.7535118680118036,
|
| 1008 |
+
"learning_rate": 6.26597584052401e-06,
|
| 1009 |
+
"loss": 0.0847,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.9568106312292359,
|
| 1014 |
+
"grad_norm": 0.7025166191190397,
|
| 1015 |
+
"learning_rate": 6.209609477998339e-06,
|
| 1016 |
+
"loss": 0.0625,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.9634551495016611,
|
| 1021 |
+
"grad_norm": 0.6972936781122409,
|
| 1022 |
+
"learning_rate": 6.153079353712201e-06,
|
| 1023 |
+
"loss": 0.1023,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.9700996677740864,
|
| 1028 |
+
"grad_norm": 0.6793230837700109,
|
| 1029 |
+
"learning_rate": 6.0963931209395165e-06,
|
| 1030 |
+
"loss": 0.0638,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.9767441860465116,
|
| 1035 |
+
"grad_norm": 0.7859561569735135,
|
| 1036 |
+
"learning_rate": 6.039558454088796e-06,
|
| 1037 |
+
"loss": 0.0921,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.9833887043189369,
|
| 1042 |
+
"grad_norm": 0.5714141172905004,
|
| 1043 |
+
"learning_rate": 5.982583047664151e-06,
|
| 1044 |
+
"loss": 0.0731,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.9900332225913622,
|
| 1049 |
+
"grad_norm": 0.7332537893766391,
|
| 1050 |
+
"learning_rate": 5.925474615223573e-06,
|
| 1051 |
+
"loss": 0.0673,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.9966777408637874,
|
| 1056 |
+
"grad_norm": 0.7492221124159332,
|
| 1057 |
+
"learning_rate": 5.8682408883346535e-06,
|
| 1058 |
+
"loss": 0.0683,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.9966777408637874,
|
| 1063 |
+
"eval_loss": 0.07248932123184204,
|
| 1064 |
+
"eval_runtime": 32.808,
|
| 1065 |
+
"eval_samples_per_second": 30.846,
|
| 1066 |
+
"eval_steps_per_second": 0.975,
|
| 1067 |
+
"step": 150
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.0033222591362125,
|
| 1071 |
+
"grad_norm": 0.869985165126094,
|
| 1072 |
+
"learning_rate": 5.810889615527839e-06,
|
| 1073 |
+
"loss": 0.0832,
|
| 1074 |
+
"step": 151
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.0099667774086378,
|
| 1078 |
+
"grad_norm": 0.7265007952878254,
|
| 1079 |
+
"learning_rate": 5.753428561247416e-06,
|
| 1080 |
+
"loss": 0.0603,
|
| 1081 |
+
"step": 152
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.0166112956810631,
|
| 1085 |
+
"grad_norm": 0.5118692934892353,
|
| 1086 |
+
"learning_rate": 5.695865504800328e-06,
|
| 1087 |
+
"loss": 0.0556,
|
| 1088 |
+
"step": 153
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.0232558139534884,
|
| 1092 |
+
"grad_norm": 0.6621056974289653,
|
| 1093 |
+
"learning_rate": 5.638208239302975e-06,
|
| 1094 |
+
"loss": 0.1026,
|
| 1095 |
+
"step": 154
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.0299003322259137,
|
| 1099 |
+
"grad_norm": 0.5532826508404691,
|
| 1100 |
+
"learning_rate": 5.5804645706261515e-06,
|
| 1101 |
+
"loss": 0.0689,
|
| 1102 |
+
"step": 155
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.0365448504983388,
|
| 1106 |
+
"grad_norm": 0.7370728155757319,
|
| 1107 |
+
"learning_rate": 5.522642316338268e-06,
|
| 1108 |
+
"loss": 0.0507,
|
| 1109 |
+
"step": 156
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.043189368770764,
|
| 1113 |
+
"grad_norm": 0.608945548724636,
|
| 1114 |
+
"learning_rate": 5.464749304646963e-06,
|
| 1115 |
+
"loss": 0.0541,
|
| 1116 |
+
"step": 157
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.0498338870431894,
|
| 1120 |
+
"grad_norm": 0.694396430915162,
|
| 1121 |
+
"learning_rate": 5.406793373339292e-06,
|
| 1122 |
+
"loss": 0.0794,
|
| 1123 |
+
"step": 158
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.0564784053156147,
|
| 1127 |
+
"grad_norm": 0.9492041254202869,
|
| 1128 |
+
"learning_rate": 5.348782368720627e-06,
|
| 1129 |
+
"loss": 0.0907,
|
| 1130 |
+
"step": 159
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.06312292358804,
|
| 1134 |
+
"grad_norm": 0.7429952070523816,
|
| 1135 |
+
"learning_rate": 5.290724144552379e-06,
|
| 1136 |
+
"loss": 0.0586,
|
| 1137 |
+
"step": 160
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.069767441860465,
|
| 1141 |
+
"grad_norm": 0.640375361353487,
|
| 1142 |
+
"learning_rate": 5.232626560988735e-06,
|
| 1143 |
+
"loss": 0.0675,
|
| 1144 |
+
"step": 161
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.0764119601328903,
|
| 1148 |
+
"grad_norm": 0.753445536719332,
|
| 1149 |
+
"learning_rate": 5.174497483512506e-06,
|
| 1150 |
+
"loss": 0.0666,
|
| 1151 |
+
"step": 162
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.0830564784053156,
|
| 1155 |
+
"grad_norm": 0.6292170840779705,
|
| 1156 |
+
"learning_rate": 5.116344781870282e-06,
|
| 1157 |
+
"loss": 0.0594,
|
| 1158 |
+
"step": 163
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.089700996677741,
|
| 1162 |
+
"grad_norm": 0.6346627170439836,
|
| 1163 |
+
"learning_rate": 5.0581763290069865e-06,
|
| 1164 |
+
"loss": 0.0523,
|
| 1165 |
+
"step": 164
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.0963455149501662,
|
| 1169 |
+
"grad_norm": 0.6561801470391864,
|
| 1170 |
+
"learning_rate": 5e-06,
|
| 1171 |
+
"loss": 0.0661,
|
| 1172 |
+
"step": 165
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.1029900332225913,
|
| 1176 |
+
"grad_norm": 0.6948146032383066,
|
| 1177 |
+
"learning_rate": 4.941823670993016e-06,
|
| 1178 |
+
"loss": 0.0706,
|
| 1179 |
+
"step": 166
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.1096345514950166,
|
| 1183 |
+
"grad_norm": 0.7613914644690916,
|
| 1184 |
+
"learning_rate": 4.883655218129719e-06,
|
| 1185 |
+
"loss": 0.0649,
|
| 1186 |
+
"step": 167
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.1162790697674418,
|
| 1190 |
+
"grad_norm": 0.45992106011709505,
|
| 1191 |
+
"learning_rate": 4.825502516487497e-06,
|
| 1192 |
+
"loss": 0.0557,
|
| 1193 |
+
"step": 168
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.1229235880398671,
|
| 1197 |
+
"grad_norm": 0.6249306843166101,
|
| 1198 |
+
"learning_rate": 4.767373439011267e-06,
|
| 1199 |
+
"loss": 0.0562,
|
| 1200 |
+
"step": 169
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.1295681063122924,
|
| 1204 |
+
"grad_norm": 0.5678874358035478,
|
| 1205 |
+
"learning_rate": 4.7092758554476215e-06,
|
| 1206 |
+
"loss": 0.0627,
|
| 1207 |
+
"step": 170
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.1362126245847177,
|
| 1211 |
+
"grad_norm": 0.37893696528451193,
|
| 1212 |
+
"learning_rate": 4.651217631279374e-06,
|
| 1213 |
+
"loss": 0.0411,
|
| 1214 |
+
"step": 171
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.1428571428571428,
|
| 1218 |
+
"grad_norm": 0.6653453617589765,
|
| 1219 |
+
"learning_rate": 4.59320662666071e-06,
|
| 1220 |
+
"loss": 0.0634,
|
| 1221 |
+
"step": 172
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.149501661129568,
|
| 1225 |
+
"grad_norm": 0.8688586555661224,
|
| 1226 |
+
"learning_rate": 4.53525069535304e-06,
|
| 1227 |
+
"loss": 0.0455,
|
| 1228 |
+
"step": 173
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.1561461794019934,
|
| 1232 |
+
"grad_norm": 0.8273055275277121,
|
| 1233 |
+
"learning_rate": 4.477357683661734e-06,
|
| 1234 |
+
"loss": 0.0732,
|
| 1235 |
+
"step": 174
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.1627906976744187,
|
| 1239 |
+
"grad_norm": 0.4300505630395071,
|
| 1240 |
+
"learning_rate": 4.4195354293738484e-06,
|
| 1241 |
+
"loss": 0.0521,
|
| 1242 |
+
"step": 175
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.169435215946844,
|
| 1246 |
+
"grad_norm": 0.4719570261098666,
|
| 1247 |
+
"learning_rate": 4.361791760697027e-06,
|
| 1248 |
+
"loss": 0.0421,
|
| 1249 |
+
"step": 176
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.176079734219269,
|
| 1253 |
+
"grad_norm": 0.7484964051921864,
|
| 1254 |
+
"learning_rate": 4.304134495199675e-06,
|
| 1255 |
+
"loss": 0.0724,
|
| 1256 |
+
"step": 177
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.1827242524916943,
|
| 1260 |
+
"grad_norm": 0.46240257112979466,
|
| 1261 |
+
"learning_rate": 4.246571438752585e-06,
|
| 1262 |
+
"loss": 0.0406,
|
| 1263 |
+
"step": 178
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.1893687707641196,
|
| 1267 |
+
"grad_norm": 0.41112129356312116,
|
| 1268 |
+
"learning_rate": 4.189110384472164e-06,
|
| 1269 |
+
"loss": 0.0538,
|
| 1270 |
+
"step": 179
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.196013289036545,
|
| 1274 |
+
"grad_norm": 0.38333151957355116,
|
| 1275 |
+
"learning_rate": 4.131759111665349e-06,
|
| 1276 |
+
"loss": 0.054,
|
| 1277 |
+
"step": 180
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.2026578073089702,
|
| 1281 |
+
"grad_norm": 0.5908193567344284,
|
| 1282 |
+
"learning_rate": 4.074525384776428e-06,
|
| 1283 |
+
"loss": 0.0731,
|
| 1284 |
+
"step": 181
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.2093023255813953,
|
| 1288 |
+
"grad_norm": 0.5085203609671776,
|
| 1289 |
+
"learning_rate": 4.017416952335849e-06,
|
| 1290 |
+
"loss": 0.0563,
|
| 1291 |
+
"step": 182
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.2159468438538206,
|
| 1295 |
+
"grad_norm": 0.6307549929539201,
|
| 1296 |
+
"learning_rate": 3.960441545911205e-06,
|
| 1297 |
+
"loss": 0.0807,
|
| 1298 |
+
"step": 183
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.2225913621262459,
|
| 1302 |
+
"grad_norm": 0.3759705322994287,
|
| 1303 |
+
"learning_rate": 3.903606879060483e-06,
|
| 1304 |
+
"loss": 0.043,
|
| 1305 |
+
"step": 184
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.2292358803986712,
|
| 1309 |
+
"grad_norm": 0.5833166358506473,
|
| 1310 |
+
"learning_rate": 3.8469206462878e-06,
|
| 1311 |
+
"loss": 0.0361,
|
| 1312 |
+
"step": 185
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.2358803986710964,
|
| 1316 |
+
"grad_norm": 0.38785628492735885,
|
| 1317 |
+
"learning_rate": 3.790390522001662e-06,
|
| 1318 |
+
"loss": 0.0394,
|
| 1319 |
+
"step": 186
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.2425249169435215,
|
| 1323 |
+
"grad_norm": 0.5384756114916416,
|
| 1324 |
+
"learning_rate": 3.7340241594759917e-06,
|
| 1325 |
+
"loss": 0.0664,
|
| 1326 |
+
"step": 187
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.2491694352159468,
|
| 1330 |
+
"grad_norm": 0.46563888733253733,
|
| 1331 |
+
"learning_rate": 3.6778291898139907e-06,
|
| 1332 |
+
"loss": 0.0488,
|
| 1333 |
+
"step": 188
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.255813953488372,
|
| 1337 |
+
"grad_norm": 0.8009617857561557,
|
| 1338 |
+
"learning_rate": 3.6218132209150047e-06,
|
| 1339 |
+
"loss": 0.0784,
|
| 1340 |
+
"step": 189
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.2624584717607974,
|
| 1344 |
+
"grad_norm": 0.5704636583609043,
|
| 1345 |
+
"learning_rate": 3.5659838364445505e-06,
|
| 1346 |
+
"loss": 0.043,
|
| 1347 |
+
"step": 190
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.2691029900332227,
|
| 1351 |
+
"grad_norm": 0.4704454938004665,
|
| 1352 |
+
"learning_rate": 3.51034859480759e-06,
|
| 1353 |
+
"loss": 0.0511,
|
| 1354 |
+
"step": 191
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.2757475083056478,
|
| 1358 |
+
"grad_norm": 0.41034160768316386,
|
| 1359 |
+
"learning_rate": 3.4549150281252635e-06,
|
| 1360 |
+
"loss": 0.0548,
|
| 1361 |
+
"step": 192
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.282392026578073,
|
| 1365 |
+
"grad_norm": 0.5724269024421317,
|
| 1366 |
+
"learning_rate": 3.399690641215142e-06,
|
| 1367 |
+
"loss": 0.0595,
|
| 1368 |
+
"step": 193
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.2890365448504983,
|
| 1372 |
+
"grad_norm": 0.5846464592698902,
|
| 1373 |
+
"learning_rate": 3.3446829105752103e-06,
|
| 1374 |
+
"loss": 0.0503,
|
| 1375 |
+
"step": 194
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.2956810631229236,
|
| 1379 |
+
"grad_norm": 0.5118811966547737,
|
| 1380 |
+
"learning_rate": 3.289899283371657e-06,
|
| 1381 |
+
"loss": 0.0524,
|
| 1382 |
+
"step": 195
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.302325581395349,
|
| 1386 |
+
"grad_norm": 0.8890139228230102,
|
| 1387 |
+
"learning_rate": 3.2353471764306567e-06,
|
| 1388 |
+
"loss": 0.0517,
|
| 1389 |
+
"step": 196
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.308970099667774,
|
| 1393 |
+
"grad_norm": 0.6318848948033396,
|
| 1394 |
+
"learning_rate": 3.1810339752342446e-06,
|
| 1395 |
+
"loss": 0.0614,
|
| 1396 |
+
"step": 197
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.3156146179401993,
|
| 1400 |
+
"grad_norm": 0.7335844136817455,
|
| 1401 |
+
"learning_rate": 3.12696703292044e-06,
|
| 1402 |
+
"loss": 0.0667,
|
| 1403 |
+
"step": 198
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.3222591362126246,
|
| 1407 |
+
"grad_norm": 0.6548135189722719,
|
| 1408 |
+
"learning_rate": 3.0731536692877596e-06,
|
| 1409 |
+
"loss": 0.0562,
|
| 1410 |
+
"step": 199
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.3289036544850499,
|
| 1414 |
+
"grad_norm": 0.46580064823387207,
|
| 1415 |
+
"learning_rate": 3.019601169804216e-06,
|
| 1416 |
+
"loss": 0.0483,
|
| 1417 |
+
"step": 200
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.3355481727574752,
|
| 1421 |
+
"grad_norm": 0.8221711324426894,
|
| 1422 |
+
"learning_rate": 2.966316784621e-06,
|
| 1423 |
+
"loss": 0.0658,
|
| 1424 |
+
"step": 201
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.3421926910299002,
|
| 1428 |
+
"grad_norm": 0.901723556867836,
|
| 1429 |
+
"learning_rate": 2.9133077275909112e-06,
|
| 1430 |
+
"loss": 0.0846,
|
| 1431 |
+
"step": 202
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.3488372093023255,
|
| 1435 |
+
"grad_norm": 0.5521866482755017,
|
| 1436 |
+
"learning_rate": 2.86058117529173e-06,
|
| 1437 |
+
"loss": 0.0516,
|
| 1438 |
+
"step": 203
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.3554817275747508,
|
| 1442 |
+
"grad_norm": 0.4587332626482284,
|
| 1443 |
+
"learning_rate": 2.8081442660546126e-06,
|
| 1444 |
+
"loss": 0.0632,
|
| 1445 |
+
"step": 204
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.3621262458471761,
|
| 1449 |
+
"grad_norm": 0.7702320166873419,
|
| 1450 |
+
"learning_rate": 2.7560040989976894e-06,
|
| 1451 |
+
"loss": 0.0908,
|
| 1452 |
+
"step": 205
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.3687707641196014,
|
| 1456 |
+
"grad_norm": 0.47296016709400723,
|
| 1457 |
+
"learning_rate": 2.7041677330649408e-06,
|
| 1458 |
+
"loss": 0.0537,
|
| 1459 |
+
"step": 206
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.3754152823920265,
|
| 1463 |
+
"grad_norm": 0.4347511906822839,
|
| 1464 |
+
"learning_rate": 2.6526421860705474e-06,
|
| 1465 |
+
"loss": 0.0471,
|
| 1466 |
+
"step": 207
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.3820598006644518,
|
| 1470 |
+
"grad_norm": 0.4808252094789194,
|
| 1471 |
+
"learning_rate": 2.601434433748771e-06,
|
| 1472 |
+
"loss": 0.0564,
|
| 1473 |
+
"step": 208
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.388704318936877,
|
| 1477 |
+
"grad_norm": 0.39354003895746387,
|
| 1478 |
+
"learning_rate": 2.550551408809566e-06,
|
| 1479 |
+
"loss": 0.0566,
|
| 1480 |
+
"step": 209
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.3953488372093024,
|
| 1484 |
+
"grad_norm": 0.4911701503722221,
|
| 1485 |
+
"learning_rate": 2.5000000000000015e-06,
|
| 1486 |
+
"loss": 0.0621,
|
| 1487 |
+
"step": 210
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.4019933554817277,
|
| 1491 |
+
"grad_norm": 0.46092436668003706,
|
| 1492 |
+
"learning_rate": 2.4497870511716237e-06,
|
| 1493 |
+
"loss": 0.0607,
|
| 1494 |
+
"step": 211
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.4086378737541527,
|
| 1498 |
+
"grad_norm": 0.5878983217004033,
|
| 1499 |
+
"learning_rate": 2.3999193603539234e-06,
|
| 1500 |
+
"loss": 0.061,
|
| 1501 |
+
"step": 212
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.415282392026578,
|
| 1505 |
+
"grad_norm": 0.522361230864803,
|
| 1506 |
+
"learning_rate": 2.3504036788339763e-06,
|
| 1507 |
+
"loss": 0.072,
|
| 1508 |
+
"step": 213
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.4219269102990033,
|
| 1512 |
+
"grad_norm": 0.4092827869499798,
|
| 1513 |
+
"learning_rate": 2.3012467102424373e-06,
|
| 1514 |
+
"loss": 0.0457,
|
| 1515 |
+
"step": 214
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.4285714285714286,
|
| 1519 |
+
"grad_norm": 0.6368421322210067,
|
| 1520 |
+
"learning_rate": 2.2524551096459703e-06,
|
| 1521 |
+
"loss": 0.0758,
|
| 1522 |
+
"step": 215
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.435215946843854,
|
| 1526 |
+
"grad_norm": 0.6403778504814651,
|
| 1527 |
+
"learning_rate": 2.204035482646267e-06,
|
| 1528 |
+
"loss": 0.0678,
|
| 1529 |
+
"step": 216
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.441860465116279,
|
| 1533 |
+
"grad_norm": 0.5188080009509758,
|
| 1534 |
+
"learning_rate": 2.155994384485742e-06,
|
| 1535 |
+
"loss": 0.0598,
|
| 1536 |
+
"step": 217
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.4485049833887043,
|
| 1540 |
+
"grad_norm": 0.48165858744860335,
|
| 1541 |
+
"learning_rate": 2.1083383191600676e-06,
|
| 1542 |
+
"loss": 0.0533,
|
| 1543 |
+
"step": 218
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.4551495016611296,
|
| 1547 |
+
"grad_norm": 0.5327724082626948,
|
| 1548 |
+
"learning_rate": 2.061073738537635e-06,
|
| 1549 |
+
"loss": 0.0522,
|
| 1550 |
+
"step": 219
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.4617940199335548,
|
| 1554 |
+
"grad_norm": 0.7299533338476814,
|
| 1555 |
+
"learning_rate": 2.0142070414860704e-06,
|
| 1556 |
+
"loss": 0.0648,
|
| 1557 |
+
"step": 220
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.4684385382059801,
|
| 1561 |
+
"grad_norm": 0.6399049060963062,
|
| 1562 |
+
"learning_rate": 1.9677445730059348e-06,
|
| 1563 |
+
"loss": 0.0646,
|
| 1564 |
+
"step": 221
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.4750830564784052,
|
| 1568 |
+
"grad_norm": 0.6611530351876729,
|
| 1569 |
+
"learning_rate": 1.9216926233717087e-06,
|
| 1570 |
+
"loss": 0.0553,
|
| 1571 |
+
"step": 222
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.4817275747508305,
|
| 1575 |
+
"grad_norm": 0.5364895163115652,
|
| 1576 |
+
"learning_rate": 1.8760574272802002e-06,
|
| 1577 |
+
"loss": 0.068,
|
| 1578 |
+
"step": 223
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.4883720930232558,
|
| 1582 |
+
"grad_norm": 0.5151753951352734,
|
| 1583 |
+
"learning_rate": 1.8308451630064484e-06,
|
| 1584 |
+
"loss": 0.0727,
|
| 1585 |
+
"step": 224
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.495016611295681,
|
| 1589 |
+
"grad_norm": 0.4311368423319901,
|
| 1590 |
+
"learning_rate": 1.7860619515673034e-06,
|
| 1591 |
+
"loss": 0.0423,
|
| 1592 |
+
"step": 225
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.5016611295681064,
|
| 1596 |
+
"grad_norm": 0.4919258897689397,
|
| 1597 |
+
"learning_rate": 1.7417138558927244e-06,
|
| 1598 |
+
"loss": 0.0405,
|
| 1599 |
+
"step": 226
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.5083056478405314,
|
| 1603 |
+
"grad_norm": 0.5521587113661647,
|
| 1604 |
+
"learning_rate": 1.6978068800049624e-06,
|
| 1605 |
+
"loss": 0.07,
|
| 1606 |
+
"step": 227
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.514950166112957,
|
| 1610 |
+
"grad_norm": 0.5044366493415394,
|
| 1611 |
+
"learning_rate": 1.6543469682057105e-06,
|
| 1612 |
+
"loss": 0.0565,
|
| 1613 |
+
"step": 228
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.521594684385382,
|
| 1617 |
+
"grad_norm": 0.39516777310730944,
|
| 1618 |
+
"learning_rate": 1.611340004271339e-06,
|
| 1619 |
+
"loss": 0.0513,
|
| 1620 |
+
"step": 229
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.5282392026578073,
|
| 1624 |
+
"grad_norm": 0.5908802744488714,
|
| 1625 |
+
"learning_rate": 1.5687918106563326e-06,
|
| 1626 |
+
"loss": 0.0505,
|
| 1627 |
+
"step": 230
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.5348837209302326,
|
| 1631 |
+
"grad_norm": 0.5993938538262349,
|
| 1632 |
+
"learning_rate": 1.5267081477050132e-06,
|
| 1633 |
+
"loss": 0.061,
|
| 1634 |
+
"step": 231
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.5415282392026577,
|
| 1638 |
+
"grad_norm": 0.5387186847707379,
|
| 1639 |
+
"learning_rate": 1.4850947128716914e-06,
|
| 1640 |
+
"loss": 0.0642,
|
| 1641 |
+
"step": 232
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.5481727574750832,
|
| 1645 |
+
"grad_norm": 0.615519559618303,
|
| 1646 |
+
"learning_rate": 1.4439571399493146e-06,
|
| 1647 |
+
"loss": 0.0547,
|
| 1648 |
+
"step": 233
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.5548172757475083,
|
| 1652 |
+
"grad_norm": 0.610572822487856,
|
| 1653 |
+
"learning_rate": 1.4033009983067454e-06,
|
| 1654 |
+
"loss": 0.0638,
|
| 1655 |
+
"step": 234
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.5614617940199336,
|
| 1659 |
+
"grad_norm": 0.48741840749475485,
|
| 1660 |
+
"learning_rate": 1.3631317921347564e-06,
|
| 1661 |
+
"loss": 0.051,
|
| 1662 |
+
"step": 235
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.5681063122923589,
|
| 1666 |
+
"grad_norm": 0.540406507264975,
|
| 1667 |
+
"learning_rate": 1.3234549597008572e-06,
|
| 1668 |
+
"loss": 0.0596,
|
| 1669 |
+
"step": 236
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.574750830564784,
|
| 1673 |
+
"grad_norm": 0.4514556860955898,
|
| 1674 |
+
"learning_rate": 1.2842758726130283e-06,
|
| 1675 |
+
"loss": 0.0562,
|
| 1676 |
+
"step": 237
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.5813953488372094,
|
| 1680 |
+
"grad_norm": 0.4493200533821844,
|
| 1681 |
+
"learning_rate": 1.2455998350925042e-06,
|
| 1682 |
+
"loss": 0.0306,
|
| 1683 |
+
"step": 238
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.5880398671096345,
|
| 1687 |
+
"grad_norm": 0.5263235492360758,
|
| 1688 |
+
"learning_rate": 1.2074320832556558e-06,
|
| 1689 |
+
"loss": 0.0602,
|
| 1690 |
+
"step": 239
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.5946843853820598,
|
| 1694 |
+
"grad_norm": 0.5740721455957148,
|
| 1695 |
+
"learning_rate": 1.1697777844051105e-06,
|
| 1696 |
+
"loss": 0.0586,
|
| 1697 |
+
"step": 240
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.601328903654485,
|
| 1701 |
+
"grad_norm": 0.6202865595174548,
|
| 1702 |
+
"learning_rate": 1.132642036330181e-06,
|
| 1703 |
+
"loss": 0.0693,
|
| 1704 |
+
"step": 241
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.6079734219269102,
|
| 1708 |
+
"grad_norm": 0.5998297967904946,
|
| 1709 |
+
"learning_rate": 1.096029866616704e-06,
|
| 1710 |
+
"loss": 0.0588,
|
| 1711 |
+
"step": 242
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.6146179401993357,
|
| 1715 |
+
"grad_norm": 0.45039052826506404,
|
| 1716 |
+
"learning_rate": 1.0599462319663906e-06,
|
| 1717 |
+
"loss": 0.0404,
|
| 1718 |
+
"step": 243
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.6212624584717608,
|
| 1722 |
+
"grad_norm": 0.45387478927844527,
|
| 1723 |
+
"learning_rate": 1.0243960175257605e-06,
|
| 1724 |
+
"loss": 0.0466,
|
| 1725 |
+
"step": 244
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.627906976744186,
|
| 1729 |
+
"grad_norm": 0.7727461157878296,
|
| 1730 |
+
"learning_rate": 9.893840362247809e-07,
|
| 1731 |
+
"loss": 0.0512,
|
| 1732 |
+
"step": 245
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.6345514950166113,
|
| 1736 |
+
"grad_norm": 0.4674812446077405,
|
| 1737 |
+
"learning_rate": 9.549150281252633e-07,
|
| 1738 |
+
"loss": 0.0621,
|
| 1739 |
+
"step": 246
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.6411960132890364,
|
| 1743 |
+
"grad_norm": 0.5112435475561715,
|
| 1744 |
+
"learning_rate": 9.209936597791407e-07,
|
| 1745 |
+
"loss": 0.0568,
|
| 1746 |
+
"step": 247
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.647840531561462,
|
| 1750 |
+
"grad_norm": 0.48326632808802955,
|
| 1751 |
+
"learning_rate": 8.876245235966884e-07,
|
| 1752 |
+
"loss": 0.0354,
|
| 1753 |
+
"step": 248
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.654485049833887,
|
| 1757 |
+
"grad_norm": 0.399153045603439,
|
| 1758 |
+
"learning_rate": 8.54812137224792e-07,
|
| 1759 |
+
"loss": 0.0462,
|
| 1760 |
+
"step": 249
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.6611295681063123,
|
| 1764 |
+
"grad_norm": 0.4550505862487184,
|
| 1765 |
+
"learning_rate": 8.225609429353187e-07,
|
| 1766 |
+
"loss": 0.0435,
|
| 1767 |
+
"step": 250
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.6677740863787376,
|
| 1771 |
+
"grad_norm": 0.5263233387777809,
|
| 1772 |
+
"learning_rate": 7.908753070237124e-07,
|
| 1773 |
+
"loss": 0.0692,
|
| 1774 |
+
"step": 251
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.6744186046511627,
|
| 1778 |
+
"grad_norm": 0.439125675201105,
|
| 1779 |
+
"learning_rate": 7.597595192178702e-07,
|
| 1780 |
+
"loss": 0.0526,
|
| 1781 |
+
"step": 252
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.6810631229235882,
|
| 1785 |
+
"grad_norm": 0.3876135531333277,
|
| 1786 |
+
"learning_rate": 7.292177920973726e-07,
|
| 1787 |
+
"loss": 0.0464,
|
| 1788 |
+
"step": 253
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 1.6877076411960132,
|
| 1792 |
+
"grad_norm": 0.3784518342351669,
|
| 1793 |
+
"learning_rate": 6.992542605231739e-07,
|
| 1794 |
+
"loss": 0.0565,
|
| 1795 |
+
"step": 254
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.6943521594684385,
|
| 1799 |
+
"grad_norm": 0.4466792986377153,
|
| 1800 |
+
"learning_rate": 6.698729810778065e-07,
|
| 1801 |
+
"loss": 0.0516,
|
| 1802 |
+
"step": 255
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 1.7009966777408638,
|
| 1806 |
+
"grad_norm": 0.75150863120178,
|
| 1807 |
+
"learning_rate": 6.410779315161885e-07,
|
| 1808 |
+
"loss": 0.065,
|
| 1809 |
+
"step": 256
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.707641196013289,
|
| 1813 |
+
"grad_norm": 0.41970908980773713,
|
| 1814 |
+
"learning_rate": 6.128730102270897e-07,
|
| 1815 |
+
"loss": 0.0403,
|
| 1816 |
+
"step": 257
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.7142857142857144,
|
| 1820 |
+
"grad_norm": 0.42761564066364777,
|
| 1821 |
+
"learning_rate": 5.852620357053651e-07,
|
| 1822 |
+
"loss": 0.0545,
|
| 1823 |
+
"step": 258
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 1.7209302325581395,
|
| 1827 |
+
"grad_norm": 0.40501611642435215,
|
| 1828 |
+
"learning_rate": 5.582487460349806e-07,
|
| 1829 |
+
"loss": 0.0426,
|
| 1830 |
+
"step": 259
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 1.7275747508305648,
|
| 1834 |
+
"grad_norm": 0.5175574150905407,
|
| 1835 |
+
"learning_rate": 5.318367983829393e-07,
|
| 1836 |
+
"loss": 0.0666,
|
| 1837 |
+
"step": 260
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 1.73421926910299,
|
| 1841 |
+
"grad_norm": 0.36925931545961604,
|
| 1842 |
+
"learning_rate": 5.06029768504166e-07,
|
| 1843 |
+
"loss": 0.0414,
|
| 1844 |
+
"step": 261
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 1.7408637873754151,
|
| 1848 |
+
"grad_norm": 0.5763761416847274,
|
| 1849 |
+
"learning_rate": 4.808311502573976e-07,
|
| 1850 |
+
"loss": 0.0638,
|
| 1851 |
+
"step": 262
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 1.7475083056478407,
|
| 1855 |
+
"grad_norm": 0.3316025388394529,
|
| 1856 |
+
"learning_rate": 4.562443551321788e-07,
|
| 1857 |
+
"loss": 0.04,
|
| 1858 |
+
"step": 263
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 1.7541528239202657,
|
| 1862 |
+
"grad_norm": 0.7139336516221676,
|
| 1863 |
+
"learning_rate": 4.322727117869951e-07,
|
| 1864 |
+
"loss": 0.0605,
|
| 1865 |
+
"step": 264
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 1.760797342192691,
|
| 1869 |
+
"grad_norm": 0.5156748893460643,
|
| 1870 |
+
"learning_rate": 4.089194655986306e-07,
|
| 1871 |
+
"loss": 0.0591,
|
| 1872 |
+
"step": 265
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 1.7674418604651163,
|
| 1876 |
+
"grad_norm": 0.7283638596556546,
|
| 1877 |
+
"learning_rate": 3.8618777822278854e-07,
|
| 1878 |
+
"loss": 0.0798,
|
| 1879 |
+
"step": 266
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 1.7740863787375414,
|
| 1883 |
+
"grad_norm": 0.5421800539182494,
|
| 1884 |
+
"learning_rate": 3.6408072716606346e-07,
|
| 1885 |
+
"loss": 0.0571,
|
| 1886 |
+
"step": 267
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 1.780730897009967,
|
| 1890 |
+
"grad_norm": 0.5090117089066726,
|
| 1891 |
+
"learning_rate": 3.426013053692878e-07,
|
| 1892 |
+
"loss": 0.058,
|
| 1893 |
+
"step": 268
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 1.787375415282392,
|
| 1897 |
+
"grad_norm": 0.49158590728514556,
|
| 1898 |
+
"learning_rate": 3.2175242080234314e-07,
|
| 1899 |
+
"loss": 0.0457,
|
| 1900 |
+
"step": 269
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 1.7940199335548173,
|
| 1904 |
+
"grad_norm": 0.4810283287867908,
|
| 1905 |
+
"learning_rate": 3.015368960704584e-07,
|
| 1906 |
+
"loss": 0.0504,
|
| 1907 |
+
"step": 270
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 1.8006644518272426,
|
| 1911 |
+
"grad_norm": 0.6267357418173619,
|
| 1912 |
+
"learning_rate": 2.819574680320825e-07,
|
| 1913 |
+
"loss": 0.0826,
|
| 1914 |
+
"step": 271
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 1.8073089700996676,
|
| 1918 |
+
"grad_norm": 0.42142389112996576,
|
| 1919 |
+
"learning_rate": 2.63016787428354e-07,
|
| 1920 |
+
"loss": 0.0432,
|
| 1921 |
+
"step": 272
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 1.8139534883720931,
|
| 1925 |
+
"grad_norm": 0.5002929091372561,
|
| 1926 |
+
"learning_rate": 2.447174185242324e-07,
|
| 1927 |
+
"loss": 0.0604,
|
| 1928 |
+
"step": 273
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 1.8205980066445182,
|
| 1932 |
+
"grad_norm": 0.42068361332599374,
|
| 1933 |
+
"learning_rate": 2.2706183876134047e-07,
|
| 1934 |
+
"loss": 0.0505,
|
| 1935 |
+
"step": 274
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 1.8272425249169435,
|
| 1939 |
+
"grad_norm": 0.4524190111477669,
|
| 1940 |
+
"learning_rate": 2.1005243842255552e-07,
|
| 1941 |
+
"loss": 0.048,
|
| 1942 |
+
"step": 275
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 1.8338870431893688,
|
| 1946 |
+
"grad_norm": 0.45609655951471145,
|
| 1947 |
+
"learning_rate": 1.9369152030840553e-07,
|
| 1948 |
+
"loss": 0.0592,
|
| 1949 |
+
"step": 276
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 1.8405315614617939,
|
| 1953 |
+
"grad_norm": 0.6678751099330384,
|
| 1954 |
+
"learning_rate": 1.779812994253055e-07,
|
| 1955 |
+
"loss": 0.0406,
|
| 1956 |
+
"step": 277
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 1.8471760797342194,
|
| 1960 |
+
"grad_norm": 0.4540634884040343,
|
| 1961 |
+
"learning_rate": 1.6292390268568103e-07,
|
| 1962 |
+
"loss": 0.0535,
|
| 1963 |
+
"step": 278
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 1.8538205980066444,
|
| 1967 |
+
"grad_norm": 0.3803108748809768,
|
| 1968 |
+
"learning_rate": 1.4852136862001766e-07,
|
| 1969 |
+
"loss": 0.0543,
|
| 1970 |
+
"step": 279
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 1.8604651162790697,
|
| 1974 |
+
"grad_norm": 1.0694423086901395,
|
| 1975 |
+
"learning_rate": 1.3477564710088097e-07,
|
| 1976 |
+
"loss": 0.0556,
|
| 1977 |
+
"step": 280
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 1.867109634551495,
|
| 1981 |
+
"grad_norm": 2.7488037869138098,
|
| 1982 |
+
"learning_rate": 1.2168859907892904e-07,
|
| 1983 |
+
"loss": 0.0941,
|
| 1984 |
+
"step": 281
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 1.87375415282392,
|
| 1988 |
+
"grad_norm": 0.5891173565335839,
|
| 1989 |
+
"learning_rate": 1.0926199633097156e-07,
|
| 1990 |
+
"loss": 0.0555,
|
| 1991 |
+
"step": 282
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 1.8803986710963456,
|
| 1995 |
+
"grad_norm": 0.41004921093494356,
|
| 1996 |
+
"learning_rate": 9.749752122010347e-08,
|
| 1997 |
+
"loss": 0.0432,
|
| 1998 |
+
"step": 283
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 1.8870431893687707,
|
| 2002 |
+
"grad_norm": 0.4128102542974997,
|
| 2003 |
+
"learning_rate": 8.639676646793382e-08,
|
| 2004 |
+
"loss": 0.0452,
|
| 2005 |
+
"step": 284
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 1.893687707641196,
|
| 2009 |
+
"grad_norm": 0.5446352118383607,
|
| 2010 |
+
"learning_rate": 7.59612349389599e-08,
|
| 2011 |
+
"loss": 0.0672,
|
| 2012 |
+
"step": 285
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 1.9003322259136213,
|
| 2016 |
+
"grad_norm": 0.5663671001216136,
|
| 2017 |
+
"learning_rate": 6.61923394371039e-08,
|
| 2018 |
+
"loss": 0.0599,
|
| 2019 |
+
"step": 286
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 1.9069767441860463,
|
| 2023 |
+
"grad_norm": 0.7575229176051346,
|
| 2024 |
+
"learning_rate": 5.709140251444201e-08,
|
| 2025 |
+
"loss": 0.0564,
|
| 2026 |
+
"step": 287
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 1.9136212624584719,
|
| 2030 |
+
"grad_norm": 0.41591234765048585,
|
| 2031 |
+
"learning_rate": 4.865965629214819e-08,
|
| 2032 |
+
"loss": 0.0582,
|
| 2033 |
+
"step": 288
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 1.920265780730897,
|
| 2037 |
+
"grad_norm": 0.3650616439731819,
|
| 2038 |
+
"learning_rate": 4.0898242293691546e-08,
|
| 2039 |
+
"loss": 0.0386,
|
| 2040 |
+
"step": 289
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 1.9269102990033222,
|
| 2044 |
+
"grad_norm": 0.6937484049329661,
|
| 2045 |
+
"learning_rate": 3.3808211290284886e-08,
|
| 2046 |
+
"loss": 0.0779,
|
| 2047 |
+
"step": 290
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 1.9335548172757475,
|
| 2051 |
+
"grad_norm": 0.3706387763514699,
|
| 2052 |
+
"learning_rate": 2.7390523158633552e-08,
|
| 2053 |
+
"loss": 0.0456,
|
| 2054 |
+
"step": 291
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 1.9401993355481728,
|
| 2058 |
+
"grad_norm": 0.40696852019057694,
|
| 2059 |
+
"learning_rate": 2.1646046750978255e-08,
|
| 2060 |
+
"loss": 0.0368,
|
| 2061 |
+
"step": 292
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 1.946843853820598,
|
| 2065 |
+
"grad_norm": 0.5221037977959876,
|
| 2066 |
+
"learning_rate": 1.657555977746972e-08,
|
| 2067 |
+
"loss": 0.0455,
|
| 2068 |
+
"step": 293
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 1.9534883720930232,
|
| 2072 |
+
"grad_norm": 0.4283204104270534,
|
| 2073 |
+
"learning_rate": 1.2179748700879013e-08,
|
| 2074 |
+
"loss": 0.0569,
|
| 2075 |
+
"step": 294
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 1.9601328903654485,
|
| 2079 |
+
"grad_norm": 0.45877963879091327,
|
| 2080 |
+
"learning_rate": 8.459208643659122e-09,
|
| 2081 |
+
"loss": 0.0523,
|
| 2082 |
+
"step": 295
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 1.9667774086378738,
|
| 2086 |
+
"grad_norm": 0.4089943640130188,
|
| 2087 |
+
"learning_rate": 5.414443307377171e-09,
|
| 2088 |
+
"loss": 0.0558,
|
| 2089 |
+
"step": 296
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 1.973421926910299,
|
| 2093 |
+
"grad_norm": 0.4565605936691547,
|
| 2094 |
+
"learning_rate": 3.0458649045211897e-09,
|
| 2095 |
+
"loss": 0.0462,
|
| 2096 |
+
"step": 297
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 1.9800664451827243,
|
| 2100 |
+
"grad_norm": 0.7146498443290045,
|
| 2101 |
+
"learning_rate": 1.3537941026914302e-09,
|
| 2102 |
+
"loss": 0.0413,
|
| 2103 |
+
"step": 298
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 1.9867109634551494,
|
| 2107 |
+
"grad_norm": 0.4588515242529337,
|
| 2108 |
+
"learning_rate": 3.384599811889766e-10,
|
| 2109 |
+
"loss": 0.0567,
|
| 2110 |
+
"step": 299
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 1.9933554817275747,
|
| 2114 |
+
"grad_norm": 0.6455401534739803,
|
| 2115 |
+
"learning_rate": 0.0,
|
| 2116 |
+
"loss": 0.0482,
|
| 2117 |
+
"step": 300
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 1.9933554817275747,
|
| 2121 |
+
"eval_loss": 0.060801174491643906,
|
| 2122 |
+
"eval_runtime": 30.1206,
|
| 2123 |
+
"eval_samples_per_second": 33.598,
|
| 2124 |
+
"eval_steps_per_second": 1.062,
|
| 2125 |
+
"step": 300
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 1.9933554817275747,
|
| 2129 |
+
"step": 300,
|
| 2130 |
+
"total_flos": 1.1631033437569024e+17,
|
| 2131 |
+
"train_loss": 0.07701069086790086,
|
| 2132 |
+
"train_runtime": 4589.0959,
|
| 2133 |
+
"train_samples_per_second": 8.375,
|
| 2134 |
+
"train_steps_per_second": 0.065
|
| 2135 |
+
}
|
| 2136 |
+
],
|
| 2137 |
+
"logging_steps": 1,
|
| 2138 |
+
"max_steps": 300,
|
| 2139 |
+
"num_input_tokens_seen": 0,
|
| 2140 |
+
"num_train_epochs": 2,
|
| 2141 |
+
"save_steps": 500,
|
| 2142 |
+
"stateful_callbacks": {
|
| 2143 |
+
"TrainerControl": {
|
| 2144 |
+
"args": {
|
| 2145 |
+
"should_epoch_stop": false,
|
| 2146 |
+
"should_evaluate": false,
|
| 2147 |
+
"should_log": false,
|
| 2148 |
+
"should_save": true,
|
| 2149 |
+
"should_training_stop": true
|
| 2150 |
+
},
|
| 2151 |
+
"attributes": {}
|
| 2152 |
+
}
|
| 2153 |
+
},
|
| 2154 |
+
"total_flos": 1.1631033437569024e+17,
|
| 2155 |
+
"train_batch_size": 8,
|
| 2156 |
+
"trial_name": null,
|
| 2157 |
+
"trial_params": null
|
| 2158 |
+
}
|