Model save
Browse files- README.md +69 -0
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +2696 -0
README.md
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: llama3
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: meta-llama/Meta-Llama-3-8B
|
| 9 |
+
datasets:
|
| 10 |
+
- generator
|
| 11 |
+
model-index:
|
| 12 |
+
- name: direct_0.1p_seed42_level1_syntax
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# direct_0.1p_seed42_level1_syntax
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 1.0811
|
| 24 |
+
|
| 25 |
+
## Model description
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Intended uses & limitations
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training and evaluation data
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training procedure
|
| 38 |
+
|
| 39 |
+
### Training hyperparameters
|
| 40 |
+
|
| 41 |
+
The following hyperparameters were used during training:
|
| 42 |
+
- learning_rate: 0.0002
|
| 43 |
+
- train_batch_size: 4
|
| 44 |
+
- eval_batch_size: 1
|
| 45 |
+
- seed: 42
|
| 46 |
+
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 4
|
| 48 |
+
- gradient_accumulation_steps: 4
|
| 49 |
+
- total_train_batch_size: 64
|
| 50 |
+
- total_eval_batch_size: 4
|
| 51 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
+
- lr_scheduler_type: cosine
|
| 53 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 54 |
+
- num_epochs: 1
|
| 55 |
+
|
| 56 |
+
### Training results
|
| 57 |
+
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 60 |
+
| 1.1069 | 0.9999 | 1888 | 1.0811 |
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
### Framework versions
|
| 64 |
+
|
| 65 |
+
- PEFT 0.11.1
|
| 66 |
+
- Transformers 4.43.4
|
| 67 |
+
- Pytorch 2.3.1+cu121
|
| 68 |
+
- Datasets 2.19.1
|
| 69 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9998676022772408,
|
| 3 |
+
"total_flos": 2.3745805538033664e+16,
|
| 4 |
+
"train_loss": 1.1182949897848953,
|
| 5 |
+
"train_runtime": 43452.0384,
|
| 6 |
+
"train_samples": 207779,
|
| 7 |
+
"train_samples_per_second": 2.781,
|
| 8 |
+
"train_steps_per_second": 0.043
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9998676022772408,
|
| 3 |
+
"total_flos": 2.3745805538033664e+16,
|
| 4 |
+
"train_loss": 1.1182949897848953,
|
| 5 |
+
"train_runtime": 43452.0384,
|
| 6 |
+
"train_samples": 207779,
|
| 7 |
+
"train_samples_per_second": 2.781,
|
| 8 |
+
"train_steps_per_second": 0.043
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2696 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9998676022772408,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1888,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0005295908910366742,
|
| 13 |
+
"grad_norm": 0.41111565141978,
|
| 14 |
+
"learning_rate": 1.0582010582010582e-06,
|
| 15 |
+
"loss": 1.3689,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0026479544551833707,
|
| 20 |
+
"grad_norm": 0.4260948383357496,
|
| 21 |
+
"learning_rate": 5.291005291005291e-06,
|
| 22 |
+
"loss": 1.3797,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.005295908910366741,
|
| 27 |
+
"grad_norm": 0.42426082631444323,
|
| 28 |
+
"learning_rate": 1.0582010582010582e-05,
|
| 29 |
+
"loss": 1.4097,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.007943863365550112,
|
| 34 |
+
"grad_norm": 0.4917431791111332,
|
| 35 |
+
"learning_rate": 1.5873015873015872e-05,
|
| 36 |
+
"loss": 1.4192,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.010591817820733483,
|
| 41 |
+
"grad_norm": 0.21890435137541137,
|
| 42 |
+
"learning_rate": 2.1164021164021164e-05,
|
| 43 |
+
"loss": 1.3564,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.013239772275916854,
|
| 48 |
+
"grad_norm": 0.1563118955046167,
|
| 49 |
+
"learning_rate": 2.6455026455026456e-05,
|
| 50 |
+
"loss": 1.3689,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.015887726731100223,
|
| 55 |
+
"grad_norm": 0.14441954891998668,
|
| 56 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 57 |
+
"loss": 1.3235,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.018535681186283594,
|
| 62 |
+
"grad_norm": 0.13273151212388007,
|
| 63 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 64 |
+
"loss": 1.2872,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.021183635641466966,
|
| 69 |
+
"grad_norm": 0.13964432905217536,
|
| 70 |
+
"learning_rate": 4.232804232804233e-05,
|
| 71 |
+
"loss": 1.2703,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.023831590096650337,
|
| 76 |
+
"grad_norm": 0.14877385709600144,
|
| 77 |
+
"learning_rate": 4.761904761904762e-05,
|
| 78 |
+
"loss": 1.286,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.026479544551833708,
|
| 83 |
+
"grad_norm": 0.11433768860303765,
|
| 84 |
+
"learning_rate": 5.291005291005291e-05,
|
| 85 |
+
"loss": 1.2649,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.02912749900701708,
|
| 90 |
+
"grad_norm": 0.09609426563043175,
|
| 91 |
+
"learning_rate": 5.82010582010582e-05,
|
| 92 |
+
"loss": 1.2586,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03177545346220045,
|
| 97 |
+
"grad_norm": 0.08894905109962493,
|
| 98 |
+
"learning_rate": 6.349206349206349e-05,
|
| 99 |
+
"loss": 1.2265,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.03442340791738382,
|
| 104 |
+
"grad_norm": 0.07490077864915894,
|
| 105 |
+
"learning_rate": 6.878306878306878e-05,
|
| 106 |
+
"loss": 1.2306,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.03707136237256719,
|
| 111 |
+
"grad_norm": 0.07227158023314931,
|
| 112 |
+
"learning_rate": 7.407407407407407e-05,
|
| 113 |
+
"loss": 1.2407,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.03971931682775056,
|
| 118 |
+
"grad_norm": 0.06869740415301212,
|
| 119 |
+
"learning_rate": 7.936507936507937e-05,
|
| 120 |
+
"loss": 1.2122,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04236727128293393,
|
| 125 |
+
"grad_norm": 0.06853429613019985,
|
| 126 |
+
"learning_rate": 8.465608465608466e-05,
|
| 127 |
+
"loss": 1.2229,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.0450152257381173,
|
| 132 |
+
"grad_norm": 0.0712797771766177,
|
| 133 |
+
"learning_rate": 8.994708994708995e-05,
|
| 134 |
+
"loss": 1.1861,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.047663180193300674,
|
| 139 |
+
"grad_norm": 0.06827620857683864,
|
| 140 |
+
"learning_rate": 9.523809523809524e-05,
|
| 141 |
+
"loss": 1.1737,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.050311134648484045,
|
| 146 |
+
"grad_norm": 0.07511648347979284,
|
| 147 |
+
"learning_rate": 0.00010052910052910055,
|
| 148 |
+
"loss": 1.1917,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.052959089103667416,
|
| 153 |
+
"grad_norm": 0.07161483769838682,
|
| 154 |
+
"learning_rate": 0.00010582010582010582,
|
| 155 |
+
"loss": 1.1657,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.05560704355885079,
|
| 160 |
+
"grad_norm": 0.0828998822809963,
|
| 161 |
+
"learning_rate": 0.00011111111111111112,
|
| 162 |
+
"loss": 1.2058,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.05825499801403416,
|
| 167 |
+
"grad_norm": 0.08432989244956403,
|
| 168 |
+
"learning_rate": 0.0001164021164021164,
|
| 169 |
+
"loss": 1.1959,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.06090295246921753,
|
| 174 |
+
"grad_norm": 0.08546417778154296,
|
| 175 |
+
"learning_rate": 0.0001216931216931217,
|
| 176 |
+
"loss": 1.1939,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.0635509069244009,
|
| 181 |
+
"grad_norm": 0.08953022960897895,
|
| 182 |
+
"learning_rate": 0.00012698412698412698,
|
| 183 |
+
"loss": 1.152,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.06619886137958426,
|
| 188 |
+
"grad_norm": 0.09090497875499474,
|
| 189 |
+
"learning_rate": 0.00013227513227513228,
|
| 190 |
+
"loss": 1.162,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.06884681583476764,
|
| 195 |
+
"grad_norm": 0.09067803395184647,
|
| 196 |
+
"learning_rate": 0.00013756613756613756,
|
| 197 |
+
"loss": 1.1564,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.071494770289951,
|
| 202 |
+
"grad_norm": 0.09746896349076282,
|
| 203 |
+
"learning_rate": 0.00014285714285714287,
|
| 204 |
+
"loss": 1.153,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.07414272474513438,
|
| 209 |
+
"grad_norm": 0.09151621194870359,
|
| 210 |
+
"learning_rate": 0.00014814814814814815,
|
| 211 |
+
"loss": 1.1733,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.07679067920031775,
|
| 216 |
+
"grad_norm": 0.09449259114304437,
|
| 217 |
+
"learning_rate": 0.00015343915343915345,
|
| 218 |
+
"loss": 1.1674,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.07943863365550112,
|
| 223 |
+
"grad_norm": 0.09545666936190501,
|
| 224 |
+
"learning_rate": 0.00015873015873015873,
|
| 225 |
+
"loss": 1.1701,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.08208658811068449,
|
| 230 |
+
"grad_norm": 0.0933818650565856,
|
| 231 |
+
"learning_rate": 0.00016402116402116404,
|
| 232 |
+
"loss": 1.1243,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.08473454256586786,
|
| 237 |
+
"grad_norm": 0.09593001870262655,
|
| 238 |
+
"learning_rate": 0.00016931216931216931,
|
| 239 |
+
"loss": 1.1347,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.08738249702105123,
|
| 244 |
+
"grad_norm": 0.09994210185656571,
|
| 245 |
+
"learning_rate": 0.00017460317460317462,
|
| 246 |
+
"loss": 1.1522,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.0900304514762346,
|
| 251 |
+
"grad_norm": 0.09791765751683923,
|
| 252 |
+
"learning_rate": 0.0001798941798941799,
|
| 253 |
+
"loss": 1.1597,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.09267840593141798,
|
| 258 |
+
"grad_norm": 0.09168024517745389,
|
| 259 |
+
"learning_rate": 0.0001851851851851852,
|
| 260 |
+
"loss": 1.1296,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.09532636038660135,
|
| 265 |
+
"grad_norm": 0.09617452615612945,
|
| 266 |
+
"learning_rate": 0.00019047619047619048,
|
| 267 |
+
"loss": 1.1261,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.09797431484178472,
|
| 272 |
+
"grad_norm": 0.10949818534072465,
|
| 273 |
+
"learning_rate": 0.0001957671957671958,
|
| 274 |
+
"loss": 1.1437,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.10062226929696809,
|
| 279 |
+
"grad_norm": 0.09788854738358516,
|
| 280 |
+
"learning_rate": 0.00019999982904458238,
|
| 281 |
+
"loss": 1.1351,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.10327022375215146,
|
| 286 |
+
"grad_norm": 0.09651211835063897,
|
| 287 |
+
"learning_rate": 0.00019999384566633964,
|
| 288 |
+
"loss": 1.1285,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.10591817820733483,
|
| 293 |
+
"grad_norm": 0.09895389323355416,
|
| 294 |
+
"learning_rate": 0.00019997931510172217,
|
| 295 |
+
"loss": 1.1592,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.1085661326625182,
|
| 300 |
+
"grad_norm": 0.09504405875958917,
|
| 301 |
+
"learning_rate": 0.0001999562385927608,
|
| 302 |
+
"loss": 1.127,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.11121408711770157,
|
| 307 |
+
"grad_norm": 0.08454569193883055,
|
| 308 |
+
"learning_rate": 0.00019992461811196922,
|
| 309 |
+
"loss": 1.1163,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.11386204157288494,
|
| 314 |
+
"grad_norm": 0.10371923765022571,
|
| 315 |
+
"learning_rate": 0.00019988445636217513,
|
| 316 |
+
"loss": 1.1286,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.11650999602806832,
|
| 321 |
+
"grad_norm": 0.09773134958204137,
|
| 322 |
+
"learning_rate": 0.00019983575677628946,
|
| 323 |
+
"loss": 1.1547,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.11915795048325169,
|
| 328 |
+
"grad_norm": 0.08853073845684854,
|
| 329 |
+
"learning_rate": 0.0001997785235170127,
|
| 330 |
+
"loss": 1.1391,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.12180590493843506,
|
| 335 |
+
"grad_norm": 0.11043850318148267,
|
| 336 |
+
"learning_rate": 0.00019971276147647937,
|
| 337 |
+
"loss": 1.1265,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.12445385939361843,
|
| 342 |
+
"grad_norm": 0.08955488246598574,
|
| 343 |
+
"learning_rate": 0.00019963847627583952,
|
| 344 |
+
"loss": 1.1293,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.1271018138488018,
|
| 349 |
+
"grad_norm": 0.08677646900814183,
|
| 350 |
+
"learning_rate": 0.00019955567426477847,
|
| 351 |
+
"loss": 1.1665,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.12974976830398516,
|
| 356 |
+
"grad_norm": 0.0827435794481087,
|
| 357 |
+
"learning_rate": 0.00019946436252097401,
|
| 358 |
+
"loss": 1.1517,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.13239772275916853,
|
| 363 |
+
"grad_norm": 0.08325869934671043,
|
| 364 |
+
"learning_rate": 0.00019936454884949146,
|
| 365 |
+
"loss": 1.1491,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1350456772143519,
|
| 370 |
+
"grad_norm": 0.08582838361774685,
|
| 371 |
+
"learning_rate": 0.0001992562417821164,
|
| 372 |
+
"loss": 1.1447,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.13769363166953527,
|
| 377 |
+
"grad_norm": 0.0784880184114551,
|
| 378 |
+
"learning_rate": 0.00019913945057662547,
|
| 379 |
+
"loss": 1.1321,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.14034158612471864,
|
| 384 |
+
"grad_norm": 0.09123702006719589,
|
| 385 |
+
"learning_rate": 0.00019901418521599506,
|
| 386 |
+
"loss": 1.0955,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.142989540579902,
|
| 391 |
+
"grad_norm": 0.08265522736674327,
|
| 392 |
+
"learning_rate": 0.000198880456407548,
|
| 393 |
+
"loss": 1.1493,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.14563749503508538,
|
| 398 |
+
"grad_norm": 0.09073002430436583,
|
| 399 |
+
"learning_rate": 0.00019873827558203826,
|
| 400 |
+
"loss": 1.1101,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.14828544949026876,
|
| 405 |
+
"grad_norm": 0.08138379527510481,
|
| 406 |
+
"learning_rate": 0.0001985876548926739,
|
| 407 |
+
"loss": 1.0982,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.15093340394545213,
|
| 412 |
+
"grad_norm": 0.09137910087915257,
|
| 413 |
+
"learning_rate": 0.00019842860721407837,
|
| 414 |
+
"loss": 1.1312,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.1535813584006355,
|
| 419 |
+
"grad_norm": 0.08477969588194346,
|
| 420 |
+
"learning_rate": 0.0001982611461411898,
|
| 421 |
+
"loss": 1.1413,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.15622931285581887,
|
| 426 |
+
"grad_norm": 0.08817787188441822,
|
| 427 |
+
"learning_rate": 0.0001980852859880992,
|
| 428 |
+
"loss": 1.1394,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.15887726731100224,
|
| 433 |
+
"grad_norm": 0.08829334171222576,
|
| 434 |
+
"learning_rate": 0.00019790104178682666,
|
| 435 |
+
"loss": 1.1236,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.1615252217661856,
|
| 440 |
+
"grad_norm": 0.09023785166170276,
|
| 441 |
+
"learning_rate": 0.00019770842928603673,
|
| 442 |
+
"loss": 1.1205,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.16417317622136898,
|
| 447 |
+
"grad_norm": 0.09500691769663268,
|
| 448 |
+
"learning_rate": 0.000197507464949692,
|
| 449 |
+
"loss": 1.1167,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.16682113067655235,
|
| 454 |
+
"grad_norm": 0.08426021874529936,
|
| 455 |
+
"learning_rate": 0.00019729816595564614,
|
| 456 |
+
"loss": 1.1106,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.16946908513173572,
|
| 461 |
+
"grad_norm": 0.08147138058421277,
|
| 462 |
+
"learning_rate": 0.00019708055019417522,
|
| 463 |
+
"loss": 1.11,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.1721170395869191,
|
| 468 |
+
"grad_norm": 0.08138544612829161,
|
| 469 |
+
"learning_rate": 0.00019685463626644873,
|
| 470 |
+
"loss": 1.1186,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.17476499404210247,
|
| 475 |
+
"grad_norm": 0.08666272634258464,
|
| 476 |
+
"learning_rate": 0.00019662044348293955,
|
| 477 |
+
"loss": 1.1372,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.17741294849728584,
|
| 482 |
+
"grad_norm": 0.0843383524929789,
|
| 483 |
+
"learning_rate": 0.00019637799186177332,
|
| 484 |
+
"loss": 1.1079,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.1800609029524692,
|
| 489 |
+
"grad_norm": 0.0772317749907282,
|
| 490 |
+
"learning_rate": 0.00019612730212701746,
|
| 491 |
+
"loss": 1.1475,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.18270885740765258,
|
| 496 |
+
"grad_norm": 0.0774927041377668,
|
| 497 |
+
"learning_rate": 0.00019586839570690957,
|
| 498 |
+
"loss": 1.1152,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.18535681186283595,
|
| 503 |
+
"grad_norm": 0.07478193615591532,
|
| 504 |
+
"learning_rate": 0.00019560129473202587,
|
| 505 |
+
"loss": 1.1267,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.18800476631801932,
|
| 510 |
+
"grad_norm": 0.08480217654240685,
|
| 511 |
+
"learning_rate": 0.00019532602203338958,
|
| 512 |
+
"loss": 1.1239,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.1906527207732027,
|
| 517 |
+
"grad_norm": 0.07940327835709149,
|
| 518 |
+
"learning_rate": 0.00019504260114051945,
|
| 519 |
+
"loss": 1.1309,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.19330067522838607,
|
| 524 |
+
"grad_norm": 0.07586072518369895,
|
| 525 |
+
"learning_rate": 0.00019475105627941828,
|
| 526 |
+
"loss": 1.134,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.19594862968356944,
|
| 531 |
+
"grad_norm": 0.07748779749674514,
|
| 532 |
+
"learning_rate": 0.00019445141237050246,
|
| 533 |
+
"loss": 1.1257,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.1985965841387528,
|
| 538 |
+
"grad_norm": 0.0749023019530035,
|
| 539 |
+
"learning_rate": 0.00019414369502647163,
|
| 540 |
+
"loss": 1.1194,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.20124453859393618,
|
| 545 |
+
"grad_norm": 0.07770003396390662,
|
| 546 |
+
"learning_rate": 0.00019382793055011942,
|
| 547 |
+
"loss": 1.1199,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.20389249304911955,
|
| 552 |
+
"grad_norm": 0.07762689051643917,
|
| 553 |
+
"learning_rate": 0.00019350414593208526,
|
| 554 |
+
"loss": 1.1427,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.20654044750430292,
|
| 559 |
+
"grad_norm": 0.07896343783224463,
|
| 560 |
+
"learning_rate": 0.00019317236884854714,
|
| 561 |
+
"loss": 1.1399,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2091884019594863,
|
| 566 |
+
"grad_norm": 0.08341099352486485,
|
| 567 |
+
"learning_rate": 0.00019283262765885614,
|
| 568 |
+
"loss": 1.0938,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.21183635641466966,
|
| 573 |
+
"grad_norm": 0.08207858561332201,
|
| 574 |
+
"learning_rate": 0.0001924849514031121,
|
| 575 |
+
"loss": 1.1024,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.21448431086985303,
|
| 580 |
+
"grad_norm": 0.07902580455460373,
|
| 581 |
+
"learning_rate": 0.00019212936979968158,
|
| 582 |
+
"loss": 1.1302,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.2171322653250364,
|
| 587 |
+
"grad_norm": 0.08126318805436292,
|
| 588 |
+
"learning_rate": 0.00019176591324265746,
|
| 589 |
+
"loss": 1.1256,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.21978021978021978,
|
| 594 |
+
"grad_norm": 0.08262896761244223,
|
| 595 |
+
"learning_rate": 0.0001913946127992611,
|
| 596 |
+
"loss": 1.1168,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.22242817423540315,
|
| 601 |
+
"grad_norm": 0.08486099971998615,
|
| 602 |
+
"learning_rate": 0.0001910155002071866,
|
| 603 |
+
"loss": 1.1093,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.22507612869058652,
|
| 608 |
+
"grad_norm": 0.08249982374177663,
|
| 609 |
+
"learning_rate": 0.0001906286078718881,
|
| 610 |
+
"loss": 1.1281,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.2277240831457699,
|
| 615 |
+
"grad_norm": 0.08241790745995113,
|
| 616 |
+
"learning_rate": 0.00019023396886380987,
|
| 617 |
+
"loss": 1.1306,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.23037203760095326,
|
| 622 |
+
"grad_norm": 0.08583672564162287,
|
| 623 |
+
"learning_rate": 0.00018983161691555947,
|
| 624 |
+
"loss": 1.1391,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.23301999205613663,
|
| 629 |
+
"grad_norm": 0.07678631371522364,
|
| 630 |
+
"learning_rate": 0.00018942158641902434,
|
| 631 |
+
"loss": 1.1249,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.23566794651132,
|
| 636 |
+
"grad_norm": 0.07687501040269233,
|
| 637 |
+
"learning_rate": 0.00018900391242243224,
|
| 638 |
+
"loss": 1.1275,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.23831590096650337,
|
| 643 |
+
"grad_norm": 0.07734964302998311,
|
| 644 |
+
"learning_rate": 0.00018857863062735525,
|
| 645 |
+
"loss": 1.1419,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.24096385542168675,
|
| 650 |
+
"grad_norm": 0.07789633231442543,
|
| 651 |
+
"learning_rate": 0.00018814577738565826,
|
| 652 |
+
"loss": 1.1313,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.24361180987687012,
|
| 657 |
+
"grad_norm": 0.0786175058826523,
|
| 658 |
+
"learning_rate": 0.00018770538969639162,
|
| 659 |
+
"loss": 1.0941,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.2462597643320535,
|
| 664 |
+
"grad_norm": 0.07662475568587293,
|
| 665 |
+
"learning_rate": 0.0001872575052026286,
|
| 666 |
+
"loss": 1.0895,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.24890771878723686,
|
| 671 |
+
"grad_norm": 0.07891566103155145,
|
| 672 |
+
"learning_rate": 0.00018680216218824763,
|
| 673 |
+
"loss": 1.1195,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.25155567324242023,
|
| 678 |
+
"grad_norm": 0.08801022195618644,
|
| 679 |
+
"learning_rate": 0.0001863393995746603,
|
| 680 |
+
"loss": 1.1123,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.2542036276976036,
|
| 685 |
+
"grad_norm": 0.07439626465552164,
|
| 686 |
+
"learning_rate": 0.00018586925691748394,
|
| 687 |
+
"loss": 1.105,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.256851582152787,
|
| 692 |
+
"grad_norm": 0.08921415332985139,
|
| 693 |
+
"learning_rate": 0.00018539177440316096,
|
| 694 |
+
"loss": 1.1183,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.2594995366079703,
|
| 699 |
+
"grad_norm": 0.0722928891718278,
|
| 700 |
+
"learning_rate": 0.00018490699284552352,
|
| 701 |
+
"loss": 1.1225,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.2621474910631537,
|
| 706 |
+
"grad_norm": 0.08123610651623102,
|
| 707 |
+
"learning_rate": 0.00018441495368230518,
|
| 708 |
+
"loss": 1.12,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.26479544551833706,
|
| 713 |
+
"grad_norm": 0.08591833713679868,
|
| 714 |
+
"learning_rate": 0.00018391569897159853,
|
| 715 |
+
"loss": 1.1164,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.26744339997352046,
|
| 720 |
+
"grad_norm": 0.0766094125907269,
|
| 721 |
+
"learning_rate": 0.0001834092713882606,
|
| 722 |
+
"loss": 1.1197,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.2700913544287038,
|
| 727 |
+
"grad_norm": 0.07548163044316222,
|
| 728 |
+
"learning_rate": 0.00018289571422026488,
|
| 729 |
+
"loss": 1.111,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.2727393088838872,
|
| 734 |
+
"grad_norm": 0.07972346954097494,
|
| 735 |
+
"learning_rate": 0.00018237507136500126,
|
| 736 |
+
"loss": 1.101,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.27538726333907054,
|
| 741 |
+
"grad_norm": 0.08328304497316567,
|
| 742 |
+
"learning_rate": 0.00018184738732552374,
|
| 743 |
+
"loss": 1.1116,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.27803521779425394,
|
| 748 |
+
"grad_norm": 0.0972393144531956,
|
| 749 |
+
"learning_rate": 0.00018131270720674658,
|
| 750 |
+
"loss": 1.1116,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.2806831722494373,
|
| 755 |
+
"grad_norm": 0.07637369852545137,
|
| 756 |
+
"learning_rate": 0.00018077107671158878,
|
| 757 |
+
"loss": 1.1091,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.2833311267046207,
|
| 762 |
+
"grad_norm": 0.07677607153425371,
|
| 763 |
+
"learning_rate": 0.00018022254213706745,
|
| 764 |
+
"loss": 1.0991,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.285979081159804,
|
| 769 |
+
"grad_norm": 0.07646177666293459,
|
| 770 |
+
"learning_rate": 0.0001796671503703406,
|
| 771 |
+
"loss": 1.1015,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.2886270356149874,
|
| 776 |
+
"grad_norm": 0.07666293082207067,
|
| 777 |
+
"learning_rate": 0.00017910494888469938,
|
| 778 |
+
"loss": 1.1007,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.29127499007017077,
|
| 783 |
+
"grad_norm": 0.075595023851119,
|
| 784 |
+
"learning_rate": 0.00017853598573550997,
|
| 785 |
+
"loss": 1.1323,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.29392294452535417,
|
| 790 |
+
"grad_norm": 0.07622954120001359,
|
| 791 |
+
"learning_rate": 0.0001779603095561062,
|
| 792 |
+
"loss": 1.1397,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.2965708989805375,
|
| 797 |
+
"grad_norm": 0.07515451588120746,
|
| 798 |
+
"learning_rate": 0.00017737796955363244,
|
| 799 |
+
"loss": 1.0916,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.2992188534357209,
|
| 804 |
+
"grad_norm": 0.07859697697366838,
|
| 805 |
+
"learning_rate": 0.0001767890155048375,
|
| 806 |
+
"loss": 1.0997,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.30186680789090425,
|
| 811 |
+
"grad_norm": 0.07702712175636178,
|
| 812 |
+
"learning_rate": 0.0001761934977518197,
|
| 813 |
+
"loss": 1.0988,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.30451476234608765,
|
| 818 |
+
"grad_norm": 0.0786046652278411,
|
| 819 |
+
"learning_rate": 0.00017559146719772417,
|
| 820 |
+
"loss": 1.1111,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.307162716801271,
|
| 825 |
+
"grad_norm": 0.07520793055259228,
|
| 826 |
+
"learning_rate": 0.00017498297530239132,
|
| 827 |
+
"loss": 1.0989,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3098106712564544,
|
| 832 |
+
"grad_norm": 0.07578073058602562,
|
| 833 |
+
"learning_rate": 0.00017436807407795852,
|
| 834 |
+
"loss": 1.0948,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.31245862571163774,
|
| 839 |
+
"grad_norm": 0.07604308595840065,
|
| 840 |
+
"learning_rate": 0.00017374681608441422,
|
| 841 |
+
"loss": 1.1036,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.31510658016682114,
|
| 846 |
+
"grad_norm": 0.08158146756884099,
|
| 847 |
+
"learning_rate": 0.0001731192544251051,
|
| 848 |
+
"loss": 1.1225,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.3177545346220045,
|
| 853 |
+
"grad_norm": 0.07498544543280497,
|
| 854 |
+
"learning_rate": 0.00017248544274219715,
|
| 855 |
+
"loss": 1.1201,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.3204024890771879,
|
| 860 |
+
"grad_norm": 0.08493466237400855,
|
| 861 |
+
"learning_rate": 0.00017184543521209036,
|
| 862 |
+
"loss": 1.103,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.3230504435323712,
|
| 867 |
+
"grad_norm": 0.08527784775049631,
|
| 868 |
+
"learning_rate": 0.00017119928654078792,
|
| 869 |
+
"loss": 1.1194,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.3256983979875546,
|
| 874 |
+
"grad_norm": 0.08169595986359099,
|
| 875 |
+
"learning_rate": 0.00017054705195922008,
|
| 876 |
+
"loss": 1.0933,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.32834635244273797,
|
| 881 |
+
"grad_norm": 0.08197785273219102,
|
| 882 |
+
"learning_rate": 0.00016988878721852334,
|
| 883 |
+
"loss": 1.1117,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.33099430689792136,
|
| 888 |
+
"grad_norm": 0.07335268515860134,
|
| 889 |
+
"learning_rate": 0.0001692245485852747,
|
| 890 |
+
"loss": 1.0865,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.3336422613531047,
|
| 895 |
+
"grad_norm": 0.08352503426226525,
|
| 896 |
+
"learning_rate": 0.00016855439283668233,
|
| 897 |
+
"loss": 1.1084,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.3362902158082881,
|
| 902 |
+
"grad_norm": 0.07593742235391308,
|
| 903 |
+
"learning_rate": 0.00016787837725573253,
|
| 904 |
+
"loss": 1.107,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.33893817026347145,
|
| 909 |
+
"grad_norm": 0.07457902202614533,
|
| 910 |
+
"learning_rate": 0.0001671965596262931,
|
| 911 |
+
"loss": 1.0935,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.34158612471865485,
|
| 916 |
+
"grad_norm": 0.08114151429238778,
|
| 917 |
+
"learning_rate": 0.00016650899822817433,
|
| 918 |
+
"loss": 1.0969,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.3442340791738382,
|
| 923 |
+
"grad_norm": 0.0717382651710646,
|
| 924 |
+
"learning_rate": 0.00016581575183214722,
|
| 925 |
+
"loss": 1.1352,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.3468820336290216,
|
| 930 |
+
"grad_norm": 0.09884940877067575,
|
| 931 |
+
"learning_rate": 0.00016511687969492024,
|
| 932 |
+
"loss": 1.0922,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.34952998808420493,
|
| 937 |
+
"grad_norm": 0.08048808120404499,
|
| 938 |
+
"learning_rate": 0.00016441244155407386,
|
| 939 |
+
"loss": 1.1225,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.35217794253938833,
|
| 944 |
+
"grad_norm": 0.07768291438253641,
|
| 945 |
+
"learning_rate": 0.00016370249762295464,
|
| 946 |
+
"loss": 1.1163,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.3548258969945717,
|
| 951 |
+
"grad_norm": 0.07669697580314543,
|
| 952 |
+
"learning_rate": 0.00016298710858552817,
|
| 953 |
+
"loss": 1.1454,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.3574738514497551,
|
| 958 |
+
"grad_norm": 0.07654243142581256,
|
| 959 |
+
"learning_rate": 0.0001622663355911922,
|
| 960 |
+
"loss": 1.0954,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.3601218059049384,
|
| 965 |
+
"grad_norm": 0.07308865070979989,
|
| 966 |
+
"learning_rate": 0.0001615402402495495,
|
| 967 |
+
"loss": 1.0849,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.3627697603601218,
|
| 972 |
+
"grad_norm": 0.07098787881731487,
|
| 973 |
+
"learning_rate": 0.00016080888462514185,
|
| 974 |
+
"loss": 1.115,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.36541771481530516,
|
| 979 |
+
"grad_norm": 0.07985291028148536,
|
| 980 |
+
"learning_rate": 0.00016007233123214486,
|
| 981 |
+
"loss": 1.0802,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.36806566927048856,
|
| 986 |
+
"grad_norm": 0.07631811342549887,
|
| 987 |
+
"learning_rate": 0.00015933064302902446,
|
| 988 |
+
"loss": 1.12,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.3707136237256719,
|
| 993 |
+
"grad_norm": 0.0716109218501906,
|
| 994 |
+
"learning_rate": 0.00015858388341315535,
|
| 995 |
+
"loss": 1.1223,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.3733615781808553,
|
| 1000 |
+
"grad_norm": 0.07770179033811983,
|
| 1001 |
+
"learning_rate": 0.00015783211621540208,
|
| 1002 |
+
"loss": 1.1206,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.37600953263603865,
|
| 1007 |
+
"grad_norm": 0.08008613710569794,
|
| 1008 |
+
"learning_rate": 0.00015707540569466282,
|
| 1009 |
+
"loss": 1.0927,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.37865748709122204,
|
| 1014 |
+
"grad_norm": 0.07909376018837275,
|
| 1015 |
+
"learning_rate": 0.00015631381653237683,
|
| 1016 |
+
"loss": 1.1208,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.3813054415464054,
|
| 1021 |
+
"grad_norm": 0.08795687346992156,
|
| 1022 |
+
"learning_rate": 0.00015554741382699556,
|
| 1023 |
+
"loss": 1.1158,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.3839533960015888,
|
| 1028 |
+
"grad_norm": 0.07837037230945529,
|
| 1029 |
+
"learning_rate": 0.00015477626308841832,
|
| 1030 |
+
"loss": 1.0996,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.38660135045677213,
|
| 1035 |
+
"grad_norm": 0.07641657124459816,
|
| 1036 |
+
"learning_rate": 0.0001540004302323927,
|
| 1037 |
+
"loss": 1.1014,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.38924930491195553,
|
| 1042 |
+
"grad_norm": 0.08137671044593779,
|
| 1043 |
+
"learning_rate": 0.00015321998157488014,
|
| 1044 |
+
"loss": 1.1078,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.3918972593671389,
|
| 1049 |
+
"grad_norm": 0.08041109732972,
|
| 1050 |
+
"learning_rate": 0.0001524349838263875,
|
| 1051 |
+
"loss": 1.1118,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.39454521382232227,
|
| 1056 |
+
"grad_norm": 0.07625661557827795,
|
| 1057 |
+
"learning_rate": 0.00015164550408626495,
|
| 1058 |
+
"loss": 1.102,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.3971931682775056,
|
| 1063 |
+
"grad_norm": 0.07622512218698127,
|
| 1064 |
+
"learning_rate": 0.00015085160983697033,
|
| 1065 |
+
"loss": 1.0988,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.399841122732689,
|
| 1070 |
+
"grad_norm": 0.07118368595229167,
|
| 1071 |
+
"learning_rate": 0.00015005336893830106,
|
| 1072 |
+
"loss": 1.1268,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.40248907718787236,
|
| 1077 |
+
"grad_norm": 0.07416345532311784,
|
| 1078 |
+
"learning_rate": 0.0001492508496215937,
|
| 1079 |
+
"loss": 1.0972,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.40513703164305576,
|
| 1084 |
+
"grad_norm": 0.07934756335393346,
|
| 1085 |
+
"learning_rate": 0.0001484441204838916,
|
| 1086 |
+
"loss": 1.1218,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.4077849860982391,
|
| 1091 |
+
"grad_norm": 0.07517515602393443,
|
| 1092 |
+
"learning_rate": 0.00014763325048208154,
|
| 1093 |
+
"loss": 1.1191,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.4104329405534225,
|
| 1098 |
+
"grad_norm": 0.07599570184456439,
|
| 1099 |
+
"learning_rate": 0.00014681830892699946,
|
| 1100 |
+
"loss": 1.1095,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.41308089500860584,
|
| 1105 |
+
"grad_norm": 0.07910162374766451,
|
| 1106 |
+
"learning_rate": 0.00014599936547750593,
|
| 1107 |
+
"loss": 1.0652,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.41572884946378924,
|
| 1112 |
+
"grad_norm": 0.07982743824380482,
|
| 1113 |
+
"learning_rate": 0.000145176490134532,
|
| 1114 |
+
"loss": 1.1065,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.4183768039189726,
|
| 1119 |
+
"grad_norm": 0.07714607042652752,
|
| 1120 |
+
"learning_rate": 0.00014434975323509566,
|
| 1121 |
+
"loss": 1.1059,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.421024758374156,
|
| 1126 |
+
"grad_norm": 0.07606594821793712,
|
| 1127 |
+
"learning_rate": 0.00014351922544628959,
|
| 1128 |
+
"loss": 1.0735,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.4236727128293393,
|
| 1133 |
+
"grad_norm": 0.07739034795460513,
|
| 1134 |
+
"learning_rate": 0.00014268497775924078,
|
| 1135 |
+
"loss": 1.1292,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4263206672845227,
|
| 1140 |
+
"grad_norm": 0.08112666266940216,
|
| 1141 |
+
"learning_rate": 0.00014184708148304255,
|
| 1142 |
+
"loss": 1.0935,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.42896862173970607,
|
| 1147 |
+
"grad_norm": 0.07417441389696992,
|
| 1148 |
+
"learning_rate": 0.000141005608238659,
|
| 1149 |
+
"loss": 1.0917,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.43161657619488947,
|
| 1154 |
+
"grad_norm": 0.07327160038369271,
|
| 1155 |
+
"learning_rate": 0.0001401606299528033,
|
| 1156 |
+
"loss": 1.1082,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.4342645306500728,
|
| 1161 |
+
"grad_norm": 0.07329251639640971,
|
| 1162 |
+
"learning_rate": 0.0001393122188517893,
|
| 1163 |
+
"loss": 1.0983,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.4369124851052562,
|
| 1168 |
+
"grad_norm": 0.07652688725710881,
|
| 1169 |
+
"learning_rate": 0.00013846044745535822,
|
| 1170 |
+
"loss": 1.1055,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.43956043956043955,
|
| 1175 |
+
"grad_norm": 0.07594879575700789,
|
| 1176 |
+
"learning_rate": 0.00013760538857047957,
|
| 1177 |
+
"loss": 1.0881,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.44220839401562295,
|
| 1182 |
+
"grad_norm": 0.07795025881048111,
|
| 1183 |
+
"learning_rate": 0.00013674711528512793,
|
| 1184 |
+
"loss": 1.135,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.4448563484708063,
|
| 1189 |
+
"grad_norm": 0.07525588786296598,
|
| 1190 |
+
"learning_rate": 0.00013588570096203554,
|
| 1191 |
+
"loss": 1.0939,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.4475043029259897,
|
| 1196 |
+
"grad_norm": 0.07793636775238774,
|
| 1197 |
+
"learning_rate": 0.00013502121923242164,
|
| 1198 |
+
"loss": 1.0912,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.45015225738117304,
|
| 1203 |
+
"grad_norm": 0.07426019016116724,
|
| 1204 |
+
"learning_rate": 0.00013415374398969844,
|
| 1205 |
+
"loss": 1.1182,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.45280021183635644,
|
| 1210 |
+
"grad_norm": 0.07684359220337984,
|
| 1211 |
+
"learning_rate": 0.00013328334938315513,
|
| 1212 |
+
"loss": 1.0972,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.4554481662915398,
|
| 1217 |
+
"grad_norm": 0.08240740529852435,
|
| 1218 |
+
"learning_rate": 0.0001324101098116196,
|
| 1219 |
+
"loss": 1.1124,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.4580961207467232,
|
| 1224 |
+
"grad_norm": 0.0763891854972936,
|
| 1225 |
+
"learning_rate": 0.00013153409991709933,
|
| 1226 |
+
"loss": 1.0883,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.4607440752019065,
|
| 1231 |
+
"grad_norm": 0.08060164757564352,
|
| 1232 |
+
"learning_rate": 0.0001306553945784009,
|
| 1233 |
+
"loss": 1.1097,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.4633920296570899,
|
| 1238 |
+
"grad_norm": 0.0769697871980839,
|
| 1239 |
+
"learning_rate": 0.00012977406890472975,
|
| 1240 |
+
"loss": 1.0881,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.46603998411227326,
|
| 1245 |
+
"grad_norm": 0.07383145111875959,
|
| 1246 |
+
"learning_rate": 0.0001288901982292701,
|
| 1247 |
+
"loss": 1.0872,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.46868793856745666,
|
| 1252 |
+
"grad_norm": 0.07293602517307703,
|
| 1253 |
+
"learning_rate": 0.0001280038581027455,
|
| 1254 |
+
"loss": 1.0712,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.47133589302264,
|
| 1259 |
+
"grad_norm": 0.07479709674097254,
|
| 1260 |
+
"learning_rate": 0.00012711512428696114,
|
| 1261 |
+
"loss": 1.0922,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.4739838474778234,
|
| 1266 |
+
"grad_norm": 0.0755945597589927,
|
| 1267 |
+
"learning_rate": 0.0001262240727483278,
|
| 1268 |
+
"loss": 1.0925,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.47663180193300675,
|
| 1273 |
+
"grad_norm": 0.07530841667815956,
|
| 1274 |
+
"learning_rate": 0.0001253307796513686,
|
| 1275 |
+
"loss": 1.1226,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.47927975638819015,
|
| 1280 |
+
"grad_norm": 0.08059154491334077,
|
| 1281 |
+
"learning_rate": 0.0001244353213522087,
|
| 1282 |
+
"loss": 1.1334,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.4819277108433735,
|
| 1287 |
+
"grad_norm": 0.07234568635036283,
|
| 1288 |
+
"learning_rate": 0.00012353777439204838,
|
| 1289 |
+
"loss": 1.1304,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.4845756652985569,
|
| 1294 |
+
"grad_norm": 0.07303744281420084,
|
| 1295 |
+
"learning_rate": 0.00012263821549062074,
|
| 1296 |
+
"loss": 1.0934,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.48722361975374023,
|
| 1301 |
+
"grad_norm": 0.07710065386256926,
|
| 1302 |
+
"learning_rate": 0.00012173672153963377,
|
| 1303 |
+
"loss": 1.1153,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.48987157420892363,
|
| 1308 |
+
"grad_norm": 0.07830816633015714,
|
| 1309 |
+
"learning_rate": 0.00012083336959619802,
|
| 1310 |
+
"loss": 1.112,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.492519528664107,
|
| 1315 |
+
"grad_norm": 0.0756022313506664,
|
| 1316 |
+
"learning_rate": 0.00011992823687623977,
|
| 1317 |
+
"loss": 1.122,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.4951674831192904,
|
| 1322 |
+
"grad_norm": 0.07163741166179009,
|
| 1323 |
+
"learning_rate": 0.00011902140074790112,
|
| 1324 |
+
"loss": 1.0669,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.4978154375744737,
|
| 1329 |
+
"grad_norm": 0.08576324173799038,
|
| 1330 |
+
"learning_rate": 0.00011811293872492644,
|
| 1331 |
+
"loss": 1.121,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5004633920296571,
|
| 1336 |
+
"grad_norm": 0.0762857316535538,
|
| 1337 |
+
"learning_rate": 0.00011720292846003706,
|
| 1338 |
+
"loss": 1.1249,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.5031113464848405,
|
| 1343 |
+
"grad_norm": 0.07839664604260674,
|
| 1344 |
+
"learning_rate": 0.00011629144773829344,
|
| 1345 |
+
"loss": 1.1062,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.5057593009400239,
|
| 1350 |
+
"grad_norm": 0.08031109603032427,
|
| 1351 |
+
"learning_rate": 0.00011537857447044656,
|
| 1352 |
+
"loss": 1.1013,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.5084072553952071,
|
| 1357 |
+
"grad_norm": 0.07756013246128472,
|
| 1358 |
+
"learning_rate": 0.00011446438668627819,
|
| 1359 |
+
"loss": 1.0908,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5110552098503905,
|
| 1364 |
+
"grad_norm": 0.07362707242031638,
|
| 1365 |
+
"learning_rate": 0.00011354896252793117,
|
| 1366 |
+
"loss": 1.0951,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.513703164305574,
|
| 1371 |
+
"grad_norm": 0.07249015180621271,
|
| 1372 |
+
"learning_rate": 0.0001126323802432301,
|
| 1373 |
+
"loss": 1.0966,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5163511187607573,
|
| 1378 |
+
"grad_norm": 0.07541567661268533,
|
| 1379 |
+
"learning_rate": 0.00011171471817899282,
|
| 1380 |
+
"loss": 1.1181,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5189990732159406,
|
| 1385 |
+
"grad_norm": 0.08071128877259519,
|
| 1386 |
+
"learning_rate": 0.0001107960547743337,
|
| 1387 |
+
"loss": 1.0811,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.521647027671124,
|
| 1392 |
+
"grad_norm": 0.08258197653732914,
|
| 1393 |
+
"learning_rate": 0.00010987646855395872,
|
| 1394 |
+
"loss": 1.1417,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5242949821263074,
|
| 1399 |
+
"grad_norm": 0.07840010957906275,
|
| 1400 |
+
"learning_rate": 0.00010895603812145359,
|
| 1401 |
+
"loss": 1.1253,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5269429365814908,
|
| 1406 |
+
"grad_norm": 0.07562413230243624,
|
| 1407 |
+
"learning_rate": 0.00010803484215256471,
|
| 1408 |
+
"loss": 1.1022,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.5295908910366741,
|
| 1413 |
+
"grad_norm": 0.0748436684540439,
|
| 1414 |
+
"learning_rate": 0.00010711295938847452,
|
| 1415 |
+
"loss": 1.0963,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.5322388454918575,
|
| 1420 |
+
"grad_norm": 0.07432957206360974,
|
| 1421 |
+
"learning_rate": 0.00010619046862907057,
|
| 1422 |
+
"loss": 1.0804,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.5348867999470409,
|
| 1427 |
+
"grad_norm": 0.07745940840059426,
|
| 1428 |
+
"learning_rate": 0.00010526744872621013,
|
| 1429 |
+
"loss": 1.1189,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.5375347544022243,
|
| 1434 |
+
"grad_norm": 0.07755117796467678,
|
| 1435 |
+
"learning_rate": 0.00010434397857698025,
|
| 1436 |
+
"loss": 1.0881,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.5401827088574076,
|
| 1441 |
+
"grad_norm": 0.07851509890246185,
|
| 1442 |
+
"learning_rate": 0.00010342013711695356,
|
| 1443 |
+
"loss": 1.1108,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.542830663312591,
|
| 1448 |
+
"grad_norm": 0.07480756214232848,
|
| 1449 |
+
"learning_rate": 0.00010249600331344144,
|
| 1450 |
+
"loss": 1.1214,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.5454786177677744,
|
| 1455 |
+
"grad_norm": 0.07388322231422537,
|
| 1456 |
+
"learning_rate": 0.00010157165615874384,
|
| 1457 |
+
"loss": 1.1129,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.5481265722229578,
|
| 1462 |
+
"grad_norm": 0.07729477706022746,
|
| 1463 |
+
"learning_rate": 0.00010064717466339746,
|
| 1464 |
+
"loss": 1.1024,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.5507745266781411,
|
| 1469 |
+
"grad_norm": 0.07418682486286104,
|
| 1470 |
+
"learning_rate": 9.972263784942191e-05,
|
| 1471 |
+
"loss": 1.0897,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.5534224811333245,
|
| 1476 |
+
"grad_norm": 0.07818304753661111,
|
| 1477 |
+
"learning_rate": 9.879812474356538e-05,
|
| 1478 |
+
"loss": 1.1209,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.5560704355885079,
|
| 1483 |
+
"grad_norm": 0.07539756214766863,
|
| 1484 |
+
"learning_rate": 9.78737143705495e-05,
|
| 1485 |
+
"loss": 1.0842,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.5587183900436913,
|
| 1490 |
+
"grad_norm": 0.08000396424013696,
|
| 1491 |
+
"learning_rate": 9.69494857463147e-05,
|
| 1492 |
+
"loss": 1.1022,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.5613663444988746,
|
| 1497 |
+
"grad_norm": 0.0780502131683453,
|
| 1498 |
+
"learning_rate": 9.602551787126593e-05,
|
| 1499 |
+
"loss": 1.0828,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.564014298954058,
|
| 1504 |
+
"grad_norm": 0.0756338343630312,
|
| 1505 |
+
"learning_rate": 9.510188972352003e-05,
|
| 1506 |
+
"loss": 1.1017,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.5666622534092414,
|
| 1511 |
+
"grad_norm": 0.0744886822174254,
|
| 1512 |
+
"learning_rate": 9.417868025215508e-05,
|
| 1513 |
+
"loss": 1.0982,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.5693102078644248,
|
| 1518 |
+
"grad_norm": 0.0743961983677683,
|
| 1519 |
+
"learning_rate": 9.325596837046177e-05,
|
| 1520 |
+
"loss": 1.1129,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.571958162319608,
|
| 1525 |
+
"grad_norm": 0.07437524331159841,
|
| 1526 |
+
"learning_rate": 9.233383294919832e-05,
|
| 1527 |
+
"loss": 1.0775,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.5746061167747915,
|
| 1532 |
+
"grad_norm": 0.0717481378509695,
|
| 1533 |
+
"learning_rate": 9.141235280984873e-05,
|
| 1534 |
+
"loss": 1.1017,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.5772540712299749,
|
| 1539 |
+
"grad_norm": 0.07661308153340447,
|
| 1540 |
+
"learning_rate": 9.04916067178854e-05,
|
| 1541 |
+
"loss": 1.0641,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.5799020256851583,
|
| 1546 |
+
"grad_norm": 0.0770623797115787,
|
| 1547 |
+
"learning_rate": 8.957167337603652e-05,
|
| 1548 |
+
"loss": 1.1182,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.5825499801403415,
|
| 1553 |
+
"grad_norm": 0.07405657746416461,
|
| 1554 |
+
"learning_rate": 8.865263141755865e-05,
|
| 1555 |
+
"loss": 1.1057,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5851979345955249,
|
| 1560 |
+
"grad_norm": 0.07499634726366083,
|
| 1561 |
+
"learning_rate": 8.773455939951548e-05,
|
| 1562 |
+
"loss": 1.091,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.5878458890507083,
|
| 1567 |
+
"grad_norm": 0.08117162442341291,
|
| 1568 |
+
"learning_rate": 8.681753579606312e-05,
|
| 1569 |
+
"loss": 1.125,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.5904938435058917,
|
| 1574 |
+
"grad_norm": 0.07499243002348535,
|
| 1575 |
+
"learning_rate": 8.590163899174202e-05,
|
| 1576 |
+
"loss": 1.0797,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.593141797961075,
|
| 1581 |
+
"grad_norm": 0.0757499666860512,
|
| 1582 |
+
"learning_rate": 8.498694727477712e-05,
|
| 1583 |
+
"loss": 1.101,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.5957897524162584,
|
| 1588 |
+
"grad_norm": 0.0790409067216223,
|
| 1589 |
+
"learning_rate": 8.40735388303861e-05,
|
| 1590 |
+
"loss": 1.1147,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.5984377068714418,
|
| 1595 |
+
"grad_norm": 0.07791231584194006,
|
| 1596 |
+
"learning_rate": 8.316149173409604e-05,
|
| 1597 |
+
"loss": 1.0985,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.6010856613266252,
|
| 1602 |
+
"grad_norm": 0.07490124978826228,
|
| 1603 |
+
"learning_rate": 8.225088394506998e-05,
|
| 1604 |
+
"loss": 1.1174,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.6037336157818085,
|
| 1609 |
+
"grad_norm": 0.07693659200902671,
|
| 1610 |
+
"learning_rate": 8.134179329944308e-05,
|
| 1611 |
+
"loss": 1.1346,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.6063815702369919,
|
| 1616 |
+
"grad_norm": 0.07605699796790699,
|
| 1617 |
+
"learning_rate": 8.043429750366941e-05,
|
| 1618 |
+
"loss": 1.1001,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.6090295246921753,
|
| 1623 |
+
"grad_norm": 0.07462322261103259,
|
| 1624 |
+
"learning_rate": 7.952847412787997e-05,
|
| 1625 |
+
"loss": 1.0907,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.6116774791473587,
|
| 1630 |
+
"grad_norm": 0.07830257859960496,
|
| 1631 |
+
"learning_rate": 7.8624400599252e-05,
|
| 1632 |
+
"loss": 1.079,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.614325433602542,
|
| 1637 |
+
"grad_norm": 0.07768781511626291,
|
| 1638 |
+
"learning_rate": 7.772215419539087e-05,
|
| 1639 |
+
"loss": 1.092,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.6169733880577254,
|
| 1644 |
+
"grad_norm": 0.07237763916654574,
|
| 1645 |
+
"learning_rate": 7.682181203772462e-05,
|
| 1646 |
+
"loss": 1.1278,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.6196213425129088,
|
| 1651 |
+
"grad_norm": 0.07421572650644807,
|
| 1652 |
+
"learning_rate": 7.592345108491177e-05,
|
| 1653 |
+
"loss": 1.0914,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.6222692969680922,
|
| 1658 |
+
"grad_norm": 0.07647986250910133,
|
| 1659 |
+
"learning_rate": 7.502714812626306e-05,
|
| 1660 |
+
"loss": 1.0973,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.6249172514232755,
|
| 1665 |
+
"grad_norm": 0.07673429125787028,
|
| 1666 |
+
"learning_rate": 7.413297977517804e-05,
|
| 1667 |
+
"loss": 1.113,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.6275652058784589,
|
| 1672 |
+
"grad_norm": 0.07847988781414174,
|
| 1673 |
+
"learning_rate": 7.324102246259588e-05,
|
| 1674 |
+
"loss": 1.1106,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.6302131603336423,
|
| 1679 |
+
"grad_norm": 0.08602822092227703,
|
| 1680 |
+
"learning_rate": 7.235135243046266e-05,
|
| 1681 |
+
"loss": 1.0951,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6328611147888257,
|
| 1686 |
+
"grad_norm": 0.07679615832799559,
|
| 1687 |
+
"learning_rate": 7.146404572521436e-05,
|
| 1688 |
+
"loss": 1.0822,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.635509069244009,
|
| 1693 |
+
"grad_norm": 0.07328313009689218,
|
| 1694 |
+
"learning_rate": 7.057917819127646e-05,
|
| 1695 |
+
"loss": 1.1144,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.6381570236991924,
|
| 1700 |
+
"grad_norm": 0.07561356041759124,
|
| 1701 |
+
"learning_rate": 6.969682546458119e-05,
|
| 1702 |
+
"loss": 1.0949,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.6408049781543758,
|
| 1707 |
+
"grad_norm": 0.07601697316878776,
|
| 1708 |
+
"learning_rate": 6.881706296610222e-05,
|
| 1709 |
+
"loss": 1.0881,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.6434529326095592,
|
| 1714 |
+
"grad_norm": 0.07424508033316624,
|
| 1715 |
+
"learning_rate": 6.793996589540794e-05,
|
| 1716 |
+
"loss": 1.0894,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.6461008870647424,
|
| 1721 |
+
"grad_norm": 0.0774103399193041,
|
| 1722 |
+
"learning_rate": 6.706560922423375e-05,
|
| 1723 |
+
"loss": 1.0889,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.6487488415199258,
|
| 1728 |
+
"grad_norm": 0.07704030882428428,
|
| 1729 |
+
"learning_rate": 6.619406769007347e-05,
|
| 1730 |
+
"loss": 1.1173,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.6513967959751092,
|
| 1735 |
+
"grad_norm": 0.07986295404405998,
|
| 1736 |
+
"learning_rate": 6.532541578979114e-05,
|
| 1737 |
+
"loss": 1.0828,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.6540447504302926,
|
| 1742 |
+
"grad_norm": 0.08205457004998916,
|
| 1743 |
+
"learning_rate": 6.445972777325322e-05,
|
| 1744 |
+
"loss": 1.1107,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.6566927048854759,
|
| 1749 |
+
"grad_norm": 0.07502810371034836,
|
| 1750 |
+
"learning_rate": 6.359707763698192e-05,
|
| 1751 |
+
"loss": 1.0947,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.6593406593406593,
|
| 1756 |
+
"grad_norm": 0.07480668675284598,
|
| 1757 |
+
"learning_rate": 6.27375391178301e-05,
|
| 1758 |
+
"loss": 1.129,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.6619886137958427,
|
| 1763 |
+
"grad_norm": 0.075324303274979,
|
| 1764 |
+
"learning_rate": 6.188118568667869e-05,
|
| 1765 |
+
"loss": 1.1049,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.6646365682510261,
|
| 1770 |
+
"grad_norm": 0.07716903325547174,
|
| 1771 |
+
"learning_rate": 6.102809054215626e-05,
|
| 1772 |
+
"loss": 1.0853,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.6672845227062094,
|
| 1777 |
+
"grad_norm": 0.08902130376105476,
|
| 1778 |
+
"learning_rate": 6.017832660438261e-05,
|
| 1779 |
+
"loss": 1.1014,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.6699324771613928,
|
| 1784 |
+
"grad_norm": 0.07427630543595487,
|
| 1785 |
+
"learning_rate": 5.9331966508735416e-05,
|
| 1786 |
+
"loss": 1.0715,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.6725804316165762,
|
| 1791 |
+
"grad_norm": 0.07483794256035971,
|
| 1792 |
+
"learning_rate": 5.848908259964176e-05,
|
| 1793 |
+
"loss": 1.1228,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.6752283860717596,
|
| 1798 |
+
"grad_norm": 0.07249249950069708,
|
| 1799 |
+
"learning_rate": 5.7649746924394374e-05,
|
| 1800 |
+
"loss": 1.1036,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.6778763405269429,
|
| 1805 |
+
"grad_norm": 0.07406719285536154,
|
| 1806 |
+
"learning_rate": 5.681403122699306e-05,
|
| 1807 |
+
"loss": 1.0975,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.6805242949821263,
|
| 1812 |
+
"grad_norm": 0.07322431567652883,
|
| 1813 |
+
"learning_rate": 5.598200694201228e-05,
|
| 1814 |
+
"loss": 1.1062,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.6831722494373097,
|
| 1819 |
+
"grad_norm": 0.07815522826559648,
|
| 1820 |
+
"learning_rate": 5.5153745188495395e-05,
|
| 1821 |
+
"loss": 1.0813,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.6858202038924931,
|
| 1826 |
+
"grad_norm": 0.07426469053250764,
|
| 1827 |
+
"learning_rate": 5.432931676387516e-05,
|
| 1828 |
+
"loss": 1.104,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.6884681583476764,
|
| 1833 |
+
"grad_norm": 0.08432084820000514,
|
| 1834 |
+
"learning_rate": 5.350879213792255e-05,
|
| 1835 |
+
"loss": 1.1244,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.6911161128028598,
|
| 1840 |
+
"grad_norm": 0.07573346123023687,
|
| 1841 |
+
"learning_rate": 5.2692241446723003e-05,
|
| 1842 |
+
"loss": 1.1084,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.6937640672580432,
|
| 1847 |
+
"grad_norm": 0.07843026074586353,
|
| 1848 |
+
"learning_rate": 5.18797344866814e-05,
|
| 1849 |
+
"loss": 1.102,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.6964120217132266,
|
| 1854 |
+
"grad_norm": 0.07870397238308142,
|
| 1855 |
+
"learning_rate": 5.107134070855617e-05,
|
| 1856 |
+
"loss": 1.0843,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.6990599761684099,
|
| 1861 |
+
"grad_norm": 0.07508751345932353,
|
| 1862 |
+
"learning_rate": 5.0267129211522834e-05,
|
| 1863 |
+
"loss": 1.0998,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.7017079306235933,
|
| 1868 |
+
"grad_norm": 0.07699932244077094,
|
| 1869 |
+
"learning_rate": 4.9467168737267376e-05,
|
| 1870 |
+
"loss": 1.1199,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.7043558850787767,
|
| 1875 |
+
"grad_norm": 0.07341861390300569,
|
| 1876 |
+
"learning_rate": 4.867152766411075e-05,
|
| 1877 |
+
"loss": 1.1099,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.7070038395339601,
|
| 1882 |
+
"grad_norm": 0.07886199590243474,
|
| 1883 |
+
"learning_rate": 4.788027400116387e-05,
|
| 1884 |
+
"loss": 1.1213,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.7096517939891434,
|
| 1889 |
+
"grad_norm": 0.07400224087388893,
|
| 1890 |
+
"learning_rate": 4.7093475382514343e-05,
|
| 1891 |
+
"loss": 1.1089,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.7122997484443268,
|
| 1896 |
+
"grad_norm": 0.07628400003826659,
|
| 1897 |
+
"learning_rate": 4.631119906144562e-05,
|
| 1898 |
+
"loss": 1.1101,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.7149477028995102,
|
| 1903 |
+
"grad_norm": 0.07655029779945569,
|
| 1904 |
+
"learning_rate": 4.5533511904687977e-05,
|
| 1905 |
+
"loss": 1.1464,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7175956573546936,
|
| 1910 |
+
"grad_norm": 0.07408511396805334,
|
| 1911 |
+
"learning_rate": 4.4760480386703085e-05,
|
| 1912 |
+
"loss": 1.1036,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.7202436118098768,
|
| 1917 |
+
"grad_norm": 0.07722020217908175,
|
| 1918 |
+
"learning_rate": 4.399217058400225e-05,
|
| 1919 |
+
"loss": 1.1229,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.7228915662650602,
|
| 1924 |
+
"grad_norm": 0.08311731202657702,
|
| 1925 |
+
"learning_rate": 4.322864816949785e-05,
|
| 1926 |
+
"loss": 1.111,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.7255395207202436,
|
| 1931 |
+
"grad_norm": 0.07589827212004248,
|
| 1932 |
+
"learning_rate": 4.2469978406890265e-05,
|
| 1933 |
+
"loss": 1.1205,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.728187475175427,
|
| 1938 |
+
"grad_norm": 0.07793858759229856,
|
| 1939 |
+
"learning_rate": 4.171622614508911e-05,
|
| 1940 |
+
"loss": 1.1104,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.7308354296306103,
|
| 1945 |
+
"grad_norm": 0.07696199883343936,
|
| 1946 |
+
"learning_rate": 4.0967455812670095e-05,
|
| 1947 |
+
"loss": 1.1055,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.7334833840857937,
|
| 1952 |
+
"grad_norm": 0.07532988662602433,
|
| 1953 |
+
"learning_rate": 4.022373141236801e-05,
|
| 1954 |
+
"loss": 1.1052,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.7361313385409771,
|
| 1959 |
+
"grad_norm": 0.07570188583228322,
|
| 1960 |
+
"learning_rate": 3.948511651560593e-05,
|
| 1961 |
+
"loss": 1.0996,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.7387792929961605,
|
| 1966 |
+
"grad_norm": 0.07840832338236073,
|
| 1967 |
+
"learning_rate": 3.87516742570611e-05,
|
| 1968 |
+
"loss": 1.0989,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.7414272474513438,
|
| 1973 |
+
"grad_norm": 0.07688842225193993,
|
| 1974 |
+
"learning_rate": 3.802346732926866e-05,
|
| 1975 |
+
"loss": 1.094,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.7440752019065272,
|
| 1980 |
+
"grad_norm": 0.07667453813475356,
|
| 1981 |
+
"learning_rate": 3.730055797726271e-05,
|
| 1982 |
+
"loss": 1.1138,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.7467231563617106,
|
| 1987 |
+
"grad_norm": 0.07256620195156938,
|
| 1988 |
+
"learning_rate": 3.658300799325568e-05,
|
| 1989 |
+
"loss": 1.096,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.749371110816894,
|
| 1994 |
+
"grad_norm": 0.07556178320717383,
|
| 1995 |
+
"learning_rate": 3.5870878711356906e-05,
|
| 1996 |
+
"loss": 1.0752,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.7520190652720773,
|
| 2001 |
+
"grad_norm": 0.0755712488652408,
|
| 2002 |
+
"learning_rate": 3.516423100232943e-05,
|
| 2003 |
+
"loss": 1.0738,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.7546670197272607,
|
| 2008 |
+
"grad_norm": 0.07976556756122673,
|
| 2009 |
+
"learning_rate": 3.446312526838726e-05,
|
| 2010 |
+
"loss": 1.0695,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.7573149741824441,
|
| 2015 |
+
"grad_norm": 0.07904402049081768,
|
| 2016 |
+
"learning_rate": 3.376762143803249e-05,
|
| 2017 |
+
"loss": 1.0979,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.7599629286376274,
|
| 2022 |
+
"grad_norm": 0.07776446642649297,
|
| 2023 |
+
"learning_rate": 3.307777896093238e-05,
|
| 2024 |
+
"loss": 1.1018,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.7626108830928108,
|
| 2029 |
+
"grad_norm": 0.07527759382185836,
|
| 2030 |
+
"learning_rate": 3.239365680283812e-05,
|
| 2031 |
+
"loss": 1.0968,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.7652588375479942,
|
| 2036 |
+
"grad_norm": 0.0776648894196293,
|
| 2037 |
+
"learning_rate": 3.171531344054448e-05,
|
| 2038 |
+
"loss": 1.0907,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.7679067920031776,
|
| 2043 |
+
"grad_norm": 0.07845291586662016,
|
| 2044 |
+
"learning_rate": 3.104280685689128e-05,
|
| 2045 |
+
"loss": 1.0988,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.7705547464583609,
|
| 2050 |
+
"grad_norm": 0.07774267939899475,
|
| 2051 |
+
"learning_rate": 3.037619453580741e-05,
|
| 2052 |
+
"loss": 1.1025,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.7732027009135443,
|
| 2057 |
+
"grad_norm": 0.07875923469354247,
|
| 2058 |
+
"learning_rate": 2.9715533457397115e-05,
|
| 2059 |
+
"loss": 1.1106,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.7758506553687277,
|
| 2064 |
+
"grad_norm": 0.07692024929981921,
|
| 2065 |
+
"learning_rate": 2.906088009306942e-05,
|
| 2066 |
+
"loss": 1.0852,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.7784986098239111,
|
| 2071 |
+
"grad_norm": 0.07785685577501977,
|
| 2072 |
+
"learning_rate": 2.841229040071145e-05,
|
| 2073 |
+
"loss": 1.0682,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.7811465642790943,
|
| 2078 |
+
"grad_norm": 0.0760319263173443,
|
| 2079 |
+
"learning_rate": 2.7769819819904876e-05,
|
| 2080 |
+
"loss": 1.1018,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.7837945187342777,
|
| 2085 |
+
"grad_norm": 0.076129060223867,
|
| 2086 |
+
"learning_rate": 2.7133523267187344e-05,
|
| 2087 |
+
"loss": 1.1027,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.7864424731894611,
|
| 2092 |
+
"grad_norm": 0.07951511310110099,
|
| 2093 |
+
"learning_rate": 2.650345513135849e-05,
|
| 2094 |
+
"loss": 1.1047,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.7890904276446445,
|
| 2099 |
+
"grad_norm": 0.0766458764221786,
|
| 2100 |
+
"learning_rate": 2.587966926883063e-05,
|
| 2101 |
+
"loss": 1.1019,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.7917383820998278,
|
| 2106 |
+
"grad_norm": 0.07749498443053499,
|
| 2107 |
+
"learning_rate": 2.526221899902539e-05,
|
| 2108 |
+
"loss": 1.1006,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.7943863365550112,
|
| 2113 |
+
"grad_norm": 0.08039092314611351,
|
| 2114 |
+
"learning_rate": 2.465115709981636e-05,
|
| 2115 |
+
"loss": 1.0787,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.7970342910101946,
|
| 2120 |
+
"grad_norm": 0.07771691438169755,
|
| 2121 |
+
"learning_rate": 2.4046535803017344e-05,
|
| 2122 |
+
"loss": 1.1049,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.799682245465378,
|
| 2127 |
+
"grad_norm": 0.07950511536169985,
|
| 2128 |
+
"learning_rate": 2.3448406789918133e-05,
|
| 2129 |
+
"loss": 1.0991,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.8023301999205613,
|
| 2134 |
+
"grad_norm": 0.08027602014187285,
|
| 2135 |
+
"learning_rate": 2.285682118686674e-05,
|
| 2136 |
+
"loss": 1.1072,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.8049781543757447,
|
| 2141 |
+
"grad_norm": 0.07696693283704584,
|
| 2142 |
+
"learning_rate": 2.227182956089924e-05,
|
| 2143 |
+
"loss": 1.0927,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.8076261088309281,
|
| 2148 |
+
"grad_norm": 0.07963101925667078,
|
| 2149 |
+
"learning_rate": 2.169348191541757e-05,
|
| 2150 |
+
"loss": 1.0934,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.8102740632861115,
|
| 2155 |
+
"grad_norm": 0.08049222177031838,
|
| 2156 |
+
"learning_rate": 2.112182768591534e-05,
|
| 2157 |
+
"loss": 1.1383,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.8129220177412948,
|
| 2162 |
+
"grad_norm": 0.07530307843428516,
|
| 2163 |
+
"learning_rate": 2.0556915735752114e-05,
|
| 2164 |
+
"loss": 1.0875,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.8155699721964782,
|
| 2169 |
+
"grad_norm": 0.0773611919685891,
|
| 2170 |
+
"learning_rate": 1.9998794351976958e-05,
|
| 2171 |
+
"loss": 1.107,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.8182179266516616,
|
| 2176 |
+
"grad_norm": 0.07662227830919144,
|
| 2177 |
+
"learning_rate": 1.9447511241200722e-05,
|
| 2178 |
+
"loss": 1.1042,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.820865881106845,
|
| 2183 |
+
"grad_norm": 0.07670312386682876,
|
| 2184 |
+
"learning_rate": 1.8903113525518334e-05,
|
| 2185 |
+
"loss": 1.0996,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.8235138355620283,
|
| 2190 |
+
"grad_norm": 0.07637261789090262,
|
| 2191 |
+
"learning_rate": 1.836564773848112e-05,
|
| 2192 |
+
"loss": 1.1056,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.8261617900172117,
|
| 2197 |
+
"grad_norm": 0.0778692527235825,
|
| 2198 |
+
"learning_rate": 1.783515982111885e-05,
|
| 2199 |
+
"loss": 1.1013,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.8288097444723951,
|
| 2204 |
+
"grad_norm": 0.07646896656564574,
|
| 2205 |
+
"learning_rate": 1.7311695118013237e-05,
|
| 2206 |
+
"loss": 1.123,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.8314576989275785,
|
| 2211 |
+
"grad_norm": 0.07814676271315506,
|
| 2212 |
+
"learning_rate": 1.6795298373421776e-05,
|
| 2213 |
+
"loss": 1.1167,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.8341056533827618,
|
| 2218 |
+
"grad_norm": 0.07623300427172139,
|
| 2219 |
+
"learning_rate": 1.6286013727453154e-05,
|
| 2220 |
+
"loss": 1.1107,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.8367536078379452,
|
| 2225 |
+
"grad_norm": 0.0770697500075822,
|
| 2226 |
+
"learning_rate": 1.5783884712294362e-05,
|
| 2227 |
+
"loss": 1.083,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.8394015622931286,
|
| 2232 |
+
"grad_norm": 0.07692133101024615,
|
| 2233 |
+
"learning_rate": 1.5288954248489685e-05,
|
| 2234 |
+
"loss": 1.1099,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.842049516748312,
|
| 2239 |
+
"grad_norm": 0.0755198202012927,
|
| 2240 |
+
"learning_rate": 1.4801264641271827e-05,
|
| 2241 |
+
"loss": 1.0753,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.8446974712034953,
|
| 2246 |
+
"grad_norm": 0.07515799406054297,
|
| 2247 |
+
"learning_rate": 1.4320857576946e-05,
|
| 2248 |
+
"loss": 1.1014,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.8473454256586787,
|
| 2253 |
+
"grad_norm": 0.07479012037004316,
|
| 2254 |
+
"learning_rate": 1.3847774119326573e-05,
|
| 2255 |
+
"loss": 1.0942,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.849993380113862,
|
| 2260 |
+
"grad_norm": 0.07721255980030046,
|
| 2261 |
+
"learning_rate": 1.3382054706226987e-05,
|
| 2262 |
+
"loss": 1.1133,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.8526413345690455,
|
| 2267 |
+
"grad_norm": 0.07642790610807988,
|
| 2268 |
+
"learning_rate": 1.2923739146003489e-05,
|
| 2269 |
+
"loss": 1.0951,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.8552892890242287,
|
| 2274 |
+
"grad_norm": 0.07855286960878372,
|
| 2275 |
+
"learning_rate": 1.2472866614152124e-05,
|
| 2276 |
+
"loss": 1.0965,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.8579372434794121,
|
| 2281 |
+
"grad_norm": 0.07681456927547702,
|
| 2282 |
+
"learning_rate": 1.2029475649960353e-05,
|
| 2283 |
+
"loss": 1.0866,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.8605851979345955,
|
| 2288 |
+
"grad_norm": 0.07738703516789706,
|
| 2289 |
+
"learning_rate": 1.1593604153212767e-05,
|
| 2290 |
+
"loss": 1.1066,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.8632331523897789,
|
| 2295 |
+
"grad_norm": 0.07595140973388262,
|
| 2296 |
+
"learning_rate": 1.1165289380951416e-05,
|
| 2297 |
+
"loss": 1.1071,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.8658811068449622,
|
| 2302 |
+
"grad_norm": 0.07583738961687224,
|
| 2303 |
+
"learning_rate": 1.074456794429135e-05,
|
| 2304 |
+
"loss": 1.1228,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.8685290613001456,
|
| 2309 |
+
"grad_norm": 0.0759554158788922,
|
| 2310 |
+
"learning_rate": 1.0331475805291136e-05,
|
| 2311 |
+
"loss": 1.1145,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.871177015755329,
|
| 2316 |
+
"grad_norm": 0.07558370365509604,
|
| 2317 |
+
"learning_rate": 9.926048273878819e-06,
|
| 2318 |
+
"loss": 1.0867,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.8738249702105124,
|
| 2323 |
+
"grad_norm": 0.07447741708164467,
|
| 2324 |
+
"learning_rate": 9.528320004833935e-06,
|
| 2325 |
+
"loss": 1.0982,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.8764729246656957,
|
| 2330 |
+
"grad_norm": 0.07741539189144307,
|
| 2331 |
+
"learning_rate": 9.138324994825186e-06,
|
| 2332 |
+
"loss": 1.0949,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.8791208791208791,
|
| 2337 |
+
"grad_norm": 0.07536616901375505,
|
| 2338 |
+
"learning_rate": 8.75609657950448e-06,
|
| 2339 |
+
"loss": 1.0941,
|
| 2340 |
+
"step": 1660
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.8817688335760625,
|
| 2344 |
+
"grad_norm": 0.07569809049947313,
|
| 2345 |
+
"learning_rate": 8.381667430657603e-06,
|
| 2346 |
+
"loss": 1.0907,
|
| 2347 |
+
"step": 1665
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.8844167880312459,
|
| 2351 |
+
"grad_norm": 0.07776927737637536,
|
| 2352 |
+
"learning_rate": 8.015069553411469e-06,
|
| 2353 |
+
"loss": 1.1134,
|
| 2354 |
+
"step": 1670
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.8870647424864292,
|
| 2358 |
+
"grad_norm": 0.07389019327060092,
|
| 2359 |
+
"learning_rate": 7.656334283498334e-06,
|
| 2360 |
+
"loss": 1.0757,
|
| 2361 |
+
"step": 1675
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.8897126969416126,
|
| 2365 |
+
"grad_norm": 0.07792778854648327,
|
| 2366 |
+
"learning_rate": 7.305492284577531e-06,
|
| 2367 |
+
"loss": 1.0795,
|
| 2368 |
+
"step": 1680
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.892360651396796,
|
| 2372 |
+
"grad_norm": 0.07609958493651112,
|
| 2373 |
+
"learning_rate": 6.962573545614159e-06,
|
| 2374 |
+
"loss": 1.0975,
|
| 2375 |
+
"step": 1685
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.8950086058519794,
|
| 2379 |
+
"grad_norm": 0.07556244044567743,
|
| 2380 |
+
"learning_rate": 6.627607378315903e-06,
|
| 2381 |
+
"loss": 1.1235,
|
| 2382 |
+
"step": 1690
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.8976565603071627,
|
| 2386 |
+
"grad_norm": 0.07801379209812083,
|
| 2387 |
+
"learning_rate": 6.3006224146274994e-06,
|
| 2388 |
+
"loss": 1.1113,
|
| 2389 |
+
"step": 1695
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.9003045147623461,
|
| 2393 |
+
"grad_norm": 0.07594137231580138,
|
| 2394 |
+
"learning_rate": 5.9816466042833085e-06,
|
| 2395 |
+
"loss": 1.1,
|
| 2396 |
+
"step": 1700
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.9029524692175295,
|
| 2400 |
+
"grad_norm": 0.0765503105906239,
|
| 2401 |
+
"learning_rate": 5.670707212418313e-06,
|
| 2402 |
+
"loss": 1.0999,
|
| 2403 |
+
"step": 1705
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.9056004236727129,
|
| 2407 |
+
"grad_norm": 0.0784364193602198,
|
| 2408 |
+
"learning_rate": 5.367830817237596e-06,
|
| 2409 |
+
"loss": 1.0789,
|
| 2410 |
+
"step": 1710
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.9082483781278962,
|
| 2414 |
+
"grad_norm": 0.07687073522520732,
|
| 2415 |
+
"learning_rate": 5.073043307744396e-06,
|
| 2416 |
+
"loss": 1.0941,
|
| 2417 |
+
"step": 1715
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.9108963325830796,
|
| 2421 |
+
"grad_norm": 0.07581921026660658,
|
| 2422 |
+
"learning_rate": 4.7863698815273195e-06,
|
| 2423 |
+
"loss": 1.1097,
|
| 2424 |
+
"step": 1720
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.913544287038263,
|
| 2428 |
+
"grad_norm": 0.07789236711738227,
|
| 2429 |
+
"learning_rate": 4.507835042606456e-06,
|
| 2430 |
+
"loss": 1.0713,
|
| 2431 |
+
"step": 1725
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.9161922414934464,
|
| 2435 |
+
"grad_norm": 0.07967848637291239,
|
| 2436 |
+
"learning_rate": 4.237462599338837e-06,
|
| 2437 |
+
"loss": 1.114,
|
| 2438 |
+
"step": 1730
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.9188401959486296,
|
| 2442 |
+
"grad_norm": 0.07753641186277775,
|
| 2443 |
+
"learning_rate": 3.975275662383404e-06,
|
| 2444 |
+
"loss": 1.086,
|
| 2445 |
+
"step": 1735
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.921488150403813,
|
| 2449 |
+
"grad_norm": 0.07788205466367802,
|
| 2450 |
+
"learning_rate": 3.721296642725569e-06,
|
| 2451 |
+
"loss": 1.1076,
|
| 2452 |
+
"step": 1740
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.9241361048589964,
|
| 2456 |
+
"grad_norm": 0.07489418836802078,
|
| 2457 |
+
"learning_rate": 3.475547249761557e-06,
|
| 2458 |
+
"loss": 1.0876,
|
| 2459 |
+
"step": 1745
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.9267840593141798,
|
| 2463 |
+
"grad_norm": 0.07633053121238625,
|
| 2464 |
+
"learning_rate": 3.23804848944278e-06,
|
| 2465 |
+
"loss": 1.1186,
|
| 2466 |
+
"step": 1750
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.9294320137693631,
|
| 2470 |
+
"grad_norm": 0.0749545234975198,
|
| 2471 |
+
"learning_rate": 3.0088206624802716e-06,
|
| 2472 |
+
"loss": 1.1176,
|
| 2473 |
+
"step": 1755
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.9320799682245465,
|
| 2477 |
+
"grad_norm": 0.07806751123666335,
|
| 2478 |
+
"learning_rate": 2.7878833626095224e-06,
|
| 2479 |
+
"loss": 1.1104,
|
| 2480 |
+
"step": 1760
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.9347279226797299,
|
| 2484 |
+
"grad_norm": 0.07350877409736523,
|
| 2485 |
+
"learning_rate": 2.575255474915561e-06,
|
| 2486 |
+
"loss": 1.1033,
|
| 2487 |
+
"step": 1765
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.9373758771349133,
|
| 2491 |
+
"grad_norm": 0.07754465442688381,
|
| 2492 |
+
"learning_rate": 2.3709551742187586e-06,
|
| 2493 |
+
"loss": 1.0918,
|
| 2494 |
+
"step": 1770
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.9400238315900966,
|
| 2498 |
+
"grad_norm": 0.07943198880463362,
|
| 2499 |
+
"learning_rate": 2.1749999235213147e-06,
|
| 2500 |
+
"loss": 1.0957,
|
| 2501 |
+
"step": 1775
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.94267178604528,
|
| 2505 |
+
"grad_norm": 0.0738980478341624,
|
| 2506 |
+
"learning_rate": 1.9874064725145746e-06,
|
| 2507 |
+
"loss": 1.0965,
|
| 2508 |
+
"step": 1780
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.9453197405004634,
|
| 2512 |
+
"grad_norm": 0.07568301411412179,
|
| 2513 |
+
"learning_rate": 1.8081908561472406e-06,
|
| 2514 |
+
"loss": 1.0928,
|
| 2515 |
+
"step": 1785
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.9479676949556468,
|
| 2519 |
+
"grad_norm": 0.07808219875357872,
|
| 2520 |
+
"learning_rate": 1.6373683932548455e-06,
|
| 2521 |
+
"loss": 1.0951,
|
| 2522 |
+
"step": 1790
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.9506156494108301,
|
| 2526 |
+
"grad_norm": 0.0768674233407858,
|
| 2527 |
+
"learning_rate": 1.4749536852502687e-06,
|
| 2528 |
+
"loss": 1.0992,
|
| 2529 |
+
"step": 1795
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.9532636038660135,
|
| 2533 |
+
"grad_norm": 0.0768177210422567,
|
| 2534 |
+
"learning_rate": 1.3209606148756993e-06,
|
| 2535 |
+
"loss": 1.1247,
|
| 2536 |
+
"step": 1800
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.9559115583211969,
|
| 2540 |
+
"grad_norm": 0.076755922231129,
|
| 2541 |
+
"learning_rate": 1.1754023450159523e-06,
|
| 2542 |
+
"loss": 1.0955,
|
| 2543 |
+
"step": 1805
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.9585595127763803,
|
| 2547 |
+
"grad_norm": 0.07587714541337397,
|
| 2548 |
+
"learning_rate": 1.0382913175733587e-06,
|
| 2549 |
+
"loss": 1.1138,
|
| 2550 |
+
"step": 1810
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.9612074672315636,
|
| 2554 |
+
"grad_norm": 0.07977716829550086,
|
| 2555 |
+
"learning_rate": 9.096392524042374e-07,
|
| 2556 |
+
"loss": 1.1018,
|
| 2557 |
+
"step": 1815
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.963855421686747,
|
| 2561 |
+
"grad_norm": 0.07805623190906948,
|
| 2562 |
+
"learning_rate": 7.894571463171851e-07,
|
| 2563 |
+
"loss": 1.0995,
|
| 2564 |
+
"step": 1820
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.9665033761419304,
|
| 2568 |
+
"grad_norm": 0.07827918723132034,
|
| 2569 |
+
"learning_rate": 6.77755272132985e-07,
|
| 2570 |
+
"loss": 1.1072,
|
| 2571 |
+
"step": 1825
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.9691513305971138,
|
| 2575 |
+
"grad_norm": 0.07504095329870522,
|
| 2576 |
+
"learning_rate": 5.745431778066412e-07,
|
| 2577 |
+
"loss": 1.0945,
|
| 2578 |
+
"step": 1830
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.9717992850522971,
|
| 2582 |
+
"grad_norm": 0.07543686944798814,
|
| 2583 |
+
"learning_rate": 4.798296856111772e-07,
|
| 2584 |
+
"loss": 1.1205,
|
| 2585 |
+
"step": 1835
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.9744472395074805,
|
| 2589 |
+
"grad_norm": 0.0775313859719524,
|
| 2590 |
+
"learning_rate": 3.9362289138351604e-07,
|
| 2591 |
+
"loss": 1.0969,
|
| 2592 |
+
"step": 1840
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.9770951939626639,
|
| 2596 |
+
"grad_norm": 0.08399677573379252,
|
| 2597 |
+
"learning_rate": 3.15930163832534e-07,
|
| 2598 |
+
"loss": 1.1326,
|
| 2599 |
+
"step": 1845
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.9797431484178473,
|
| 2603 |
+
"grad_norm": 0.08167978103101492,
|
| 2604 |
+
"learning_rate": 2.4675814390916483e-07,
|
| 2605 |
+
"loss": 1.1006,
|
| 2606 |
+
"step": 1850
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.9823911028730306,
|
| 2610 |
+
"grad_norm": 0.07627980254031101,
|
| 2611 |
+
"learning_rate": 1.8611274423874226e-07,
|
| 2612 |
+
"loss": 1.1026,
|
| 2613 |
+
"step": 1855
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.985039057328214,
|
| 2617 |
+
"grad_norm": 0.07798059783576732,
|
| 2618 |
+
"learning_rate": 1.339991486156489e-07,
|
| 2619 |
+
"loss": 1.112,
|
| 2620 |
+
"step": 1860
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.9876870117833974,
|
| 2624 |
+
"grad_norm": 0.07847559734854997,
|
| 2625 |
+
"learning_rate": 9.04218115601485e-08,
|
| 2626 |
+
"loss": 1.0969,
|
| 2627 |
+
"step": 1865
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.9903349662385807,
|
| 2631 |
+
"grad_norm": 0.07602247898283418,
|
| 2632 |
+
"learning_rate": 5.5384457937723843e-08,
|
| 2633 |
+
"loss": 1.0857,
|
| 2634 |
+
"step": 1870
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.992982920693764,
|
| 2638 |
+
"grad_norm": 0.07830104923923938,
|
| 2639 |
+
"learning_rate": 2.8890082640598004e-08,
|
| 2640 |
+
"loss": 1.0924,
|
| 2641 |
+
"step": 1875
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.9956308751489474,
|
| 2645 |
+
"grad_norm": 0.0737777189572401,
|
| 2646 |
+
"learning_rate": 1.0940950331772559e-08,
|
| 2647 |
+
"loss": 1.0997,
|
| 2648 |
+
"step": 1880
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.9982788296041308,
|
| 2652 |
+
"grad_norm": 0.07376783137923695,
|
| 2653 |
+
"learning_rate": 1.5385952514712644e-09,
|
| 2654 |
+
"loss": 1.1069,
|
| 2655 |
+
"step": 1885
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.9998676022772408,
|
| 2659 |
+
"eval_loss": 1.0810853242874146,
|
| 2660 |
+
"eval_runtime": 4476.4649,
|
| 2661 |
+
"eval_samples_per_second": 3.006,
|
| 2662 |
+
"eval_steps_per_second": 0.752,
|
| 2663 |
+
"step": 1888
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 0.9998676022772408,
|
| 2667 |
+
"step": 1888,
|
| 2668 |
+
"total_flos": 2.3745805538033664e+16,
|
| 2669 |
+
"train_loss": 1.1182949897848953,
|
| 2670 |
+
"train_runtime": 43452.0384,
|
| 2671 |
+
"train_samples_per_second": 2.781,
|
| 2672 |
+
"train_steps_per_second": 0.043
|
| 2673 |
+
}
|
| 2674 |
+
],
|
| 2675 |
+
"logging_steps": 5,
|
| 2676 |
+
"max_steps": 1888,
|
| 2677 |
+
"num_input_tokens_seen": 0,
|
| 2678 |
+
"num_train_epochs": 1,
|
| 2679 |
+
"save_steps": 100,
|
| 2680 |
+
"stateful_callbacks": {
|
| 2681 |
+
"TrainerControl": {
|
| 2682 |
+
"args": {
|
| 2683 |
+
"should_epoch_stop": false,
|
| 2684 |
+
"should_evaluate": false,
|
| 2685 |
+
"should_log": false,
|
| 2686 |
+
"should_save": true,
|
| 2687 |
+
"should_training_stop": true
|
| 2688 |
+
},
|
| 2689 |
+
"attributes": {}
|
| 2690 |
+
}
|
| 2691 |
+
},
|
| 2692 |
+
"total_flos": 2.3745805538033664e+16,
|
| 2693 |
+
"train_batch_size": 4,
|
| 2694 |
+
"trial_name": null,
|
| 2695 |
+
"trial_params": null
|
| 2696 |
+
}
|