Model save
Browse files- README.md +73 -0
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +2310 -0
README.md
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: meta-llama/Meta-Llama-3-8B-Instruct
|
| 3 |
+
library_name: peft
|
| 4 |
+
license: llama3
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: Meta-Llama-3-8B-Instruct-mirage-mirage-gpt-4o-sft-instruct-llama-3
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# Meta-Llama-3-8B-Instruct-mirage-mirage-gpt-4o-sft-instruct-llama-3
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.2510
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0002
|
| 41 |
+
- train_batch_size: 2
|
| 42 |
+
- eval_batch_size: 2
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 4
|
| 46 |
+
- gradient_accumulation_steps: 2
|
| 47 |
+
- total_train_batch_size: 16
|
| 48 |
+
- total_eval_batch_size: 8
|
| 49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 50 |
+
- lr_scheduler_type: cosine
|
| 51 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 52 |
+
- num_epochs: 1
|
| 53 |
+
|
| 54 |
+
### Training results
|
| 55 |
+
|
| 56 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 57 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 58 |
+
| 0.3599 | 0.1268 | 200 | 0.3172 |
|
| 59 |
+
| 0.3295 | 0.2536 | 400 | 0.2919 |
|
| 60 |
+
| 0.323 | 0.3803 | 600 | 0.2789 |
|
| 61 |
+
| 0.3274 | 0.5071 | 800 | 0.2686 |
|
| 62 |
+
| 0.3171 | 0.6339 | 1000 | 0.2597 |
|
| 63 |
+
| 0.3034 | 0.7607 | 1200 | 0.2540 |
|
| 64 |
+
| 0.265 | 0.8875 | 1400 | 0.2510 |
|
| 65 |
+
|
| 66 |
+
|
| 67 |
+
### Framework versions
|
| 68 |
+
|
| 69 |
+
- PEFT 0.10.0
|
| 70 |
+
- Transformers 4.44.0
|
| 71 |
+
- Pytorch 2.4.0+cu121
|
| 72 |
+
- Datasets 2.20.0
|
| 73 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9996830427892235,
|
| 3 |
+
"total_flos": 8013042675351552.0,
|
| 4 |
+
"train_loss": 0.3010184336819827,
|
| 5 |
+
"train_runtime": 18281.3669,
|
| 6 |
+
"train_samples": 25239,
|
| 7 |
+
"train_samples_per_second": 1.381,
|
| 8 |
+
"train_steps_per_second": 0.086
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9996830427892235,
|
| 3 |
+
"total_flos": 8013042675351552.0,
|
| 4 |
+
"train_loss": 0.3010184336819827,
|
| 5 |
+
"train_runtime": 18281.3669,
|
| 6 |
+
"train_samples": 25239,
|
| 7 |
+
"train_samples_per_second": 1.381,
|
| 8 |
+
"train_steps_per_second": 0.086
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2310 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9996830427892235,
|
| 5 |
+
"eval_steps": 200,
|
| 6 |
+
"global_step": 1577,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0006339144215530904,
|
| 13 |
+
"grad_norm": 0.1844951284226507,
|
| 14 |
+
"learning_rate": 1.2658227848101265e-06,
|
| 15 |
+
"loss": 0.2445,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.003169572107765452,
|
| 20 |
+
"grad_norm": 0.4228474323435137,
|
| 21 |
+
"learning_rate": 6.329113924050633e-06,
|
| 22 |
+
"loss": 0.5751,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.006339144215530904,
|
| 27 |
+
"grad_norm": 0.44943947868261674,
|
| 28 |
+
"learning_rate": 1.2658227848101267e-05,
|
| 29 |
+
"loss": 0.6355,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.009508716323296355,
|
| 34 |
+
"grad_norm": 0.4388677072263688,
|
| 35 |
+
"learning_rate": 1.89873417721519e-05,
|
| 36 |
+
"loss": 0.5644,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.012678288431061807,
|
| 41 |
+
"grad_norm": 0.42271873762000256,
|
| 42 |
+
"learning_rate": 2.5316455696202533e-05,
|
| 43 |
+
"loss": 0.6074,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.01584786053882726,
|
| 48 |
+
"grad_norm": 0.394168038759961,
|
| 49 |
+
"learning_rate": 3.1645569620253167e-05,
|
| 50 |
+
"loss": 0.5618,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.01901743264659271,
|
| 55 |
+
"grad_norm": 0.3572253194331872,
|
| 56 |
+
"learning_rate": 3.79746835443038e-05,
|
| 57 |
+
"loss": 0.3782,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.022187004754358162,
|
| 62 |
+
"grad_norm": 0.412555556410598,
|
| 63 |
+
"learning_rate": 4.430379746835443e-05,
|
| 64 |
+
"loss": 0.5004,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.025356576862123614,
|
| 69 |
+
"grad_norm": 0.3794217410787016,
|
| 70 |
+
"learning_rate": 5.0632911392405066e-05,
|
| 71 |
+
"loss": 0.4466,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.028526148969889066,
|
| 76 |
+
"grad_norm": 0.35048440162730615,
|
| 77 |
+
"learning_rate": 5.69620253164557e-05,
|
| 78 |
+
"loss": 0.4673,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.03169572107765452,
|
| 83 |
+
"grad_norm": 0.5944470916433041,
|
| 84 |
+
"learning_rate": 6.329113924050633e-05,
|
| 85 |
+
"loss": 0.4321,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.03486529318541997,
|
| 90 |
+
"grad_norm": 0.42565747209046223,
|
| 91 |
+
"learning_rate": 6.962025316455697e-05,
|
| 92 |
+
"loss": 0.3458,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03803486529318542,
|
| 97 |
+
"grad_norm": 0.4086754614850085,
|
| 98 |
+
"learning_rate": 7.59493670886076e-05,
|
| 99 |
+
"loss": 0.3853,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.04120443740095087,
|
| 104 |
+
"grad_norm": 0.4100402315211435,
|
| 105 |
+
"learning_rate": 8.227848101265824e-05,
|
| 106 |
+
"loss": 0.3248,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.044374009508716325,
|
| 111 |
+
"grad_norm": 0.439691902474589,
|
| 112 |
+
"learning_rate": 8.860759493670887e-05,
|
| 113 |
+
"loss": 0.3902,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04754358161648178,
|
| 118 |
+
"grad_norm": 0.4789983229510449,
|
| 119 |
+
"learning_rate": 9.493670886075949e-05,
|
| 120 |
+
"loss": 0.4375,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.05071315372424723,
|
| 125 |
+
"grad_norm": 0.4326295423185557,
|
| 126 |
+
"learning_rate": 0.00010126582278481013,
|
| 127 |
+
"loss": 0.344,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.05388272583201268,
|
| 132 |
+
"grad_norm": 0.4082745764246507,
|
| 133 |
+
"learning_rate": 0.00010759493670886076,
|
| 134 |
+
"loss": 0.373,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.05705229793977813,
|
| 139 |
+
"grad_norm": 0.44917869809554406,
|
| 140 |
+
"learning_rate": 0.0001139240506329114,
|
| 141 |
+
"loss": 0.3366,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.060221870047543584,
|
| 146 |
+
"grad_norm": 0.4099898854278625,
|
| 147 |
+
"learning_rate": 0.00012025316455696203,
|
| 148 |
+
"loss": 0.3827,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.06339144215530904,
|
| 153 |
+
"grad_norm": 0.5173612120396457,
|
| 154 |
+
"learning_rate": 0.00012658227848101267,
|
| 155 |
+
"loss": 0.3913,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.06656101426307448,
|
| 160 |
+
"grad_norm": 0.4695908910723305,
|
| 161 |
+
"learning_rate": 0.0001329113924050633,
|
| 162 |
+
"loss": 0.3285,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.06973058637083994,
|
| 167 |
+
"grad_norm": 0.34610029250066504,
|
| 168 |
+
"learning_rate": 0.00013924050632911395,
|
| 169 |
+
"loss": 0.3542,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.07290015847860538,
|
| 174 |
+
"grad_norm": 0.3833563232036365,
|
| 175 |
+
"learning_rate": 0.00014556962025316457,
|
| 176 |
+
"loss": 0.3442,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.07606973058637084,
|
| 181 |
+
"grad_norm": 0.38597736664868315,
|
| 182 |
+
"learning_rate": 0.0001518987341772152,
|
| 183 |
+
"loss": 0.3499,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.07923930269413629,
|
| 188 |
+
"grad_norm": 0.4555824320300245,
|
| 189 |
+
"learning_rate": 0.00015822784810126583,
|
| 190 |
+
"loss": 0.3843,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.08240887480190175,
|
| 195 |
+
"grad_norm": 0.44058959604469733,
|
| 196 |
+
"learning_rate": 0.00016455696202531648,
|
| 197 |
+
"loss": 0.3321,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.08557844690966719,
|
| 202 |
+
"grad_norm": 0.37513672150754146,
|
| 203 |
+
"learning_rate": 0.0001708860759493671,
|
| 204 |
+
"loss": 0.3409,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.08874801901743265,
|
| 209 |
+
"grad_norm": 0.3532888739409051,
|
| 210 |
+
"learning_rate": 0.00017721518987341773,
|
| 211 |
+
"loss": 0.3388,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.0919175911251981,
|
| 216 |
+
"grad_norm": 0.31398944959900404,
|
| 217 |
+
"learning_rate": 0.00018354430379746836,
|
| 218 |
+
"loss": 0.3407,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.09508716323296355,
|
| 223 |
+
"grad_norm": 0.48473648286443866,
|
| 224 |
+
"learning_rate": 0.00018987341772151899,
|
| 225 |
+
"loss": 0.4109,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.098256735340729,
|
| 230 |
+
"grad_norm": 0.3832743712760423,
|
| 231 |
+
"learning_rate": 0.00019620253164556964,
|
| 232 |
+
"loss": 0.2894,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.10142630744849446,
|
| 237 |
+
"grad_norm": 0.3576599310136604,
|
| 238 |
+
"learning_rate": 0.00019999901968817678,
|
| 239 |
+
"loss": 0.3685,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.1045958795562599,
|
| 244 |
+
"grad_norm": 0.4041268184733326,
|
| 245 |
+
"learning_rate": 0.0001999879914008964,
|
| 246 |
+
"loss": 0.3103,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.10776545166402536,
|
| 251 |
+
"grad_norm": 0.348710082889974,
|
| 252 |
+
"learning_rate": 0.00019996471079244477,
|
| 253 |
+
"loss": 0.3686,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.1109350237717908,
|
| 258 |
+
"grad_norm": 0.3641139077278622,
|
| 259 |
+
"learning_rate": 0.0001999291807155794,
|
| 260 |
+
"loss": 0.3672,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.11410459587955626,
|
| 265 |
+
"grad_norm": 0.34875291735749603,
|
| 266 |
+
"learning_rate": 0.0001998814055240823,
|
| 267 |
+
"loss": 0.3289,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.11727416798732171,
|
| 272 |
+
"grad_norm": 0.35868082118594846,
|
| 273 |
+
"learning_rate": 0.00019982139107222632,
|
| 274 |
+
"loss": 0.3843,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.12044374009508717,
|
| 279 |
+
"grad_norm": 0.2975053354861811,
|
| 280 |
+
"learning_rate": 0.000199749144714058,
|
| 281 |
+
"loss": 0.3187,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.12361331220285261,
|
| 286 |
+
"grad_norm": 0.3926097041806586,
|
| 287 |
+
"learning_rate": 0.00019966467530249627,
|
| 288 |
+
"loss": 0.3711,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.12678288431061807,
|
| 293 |
+
"grad_norm": 0.39235636818547276,
|
| 294 |
+
"learning_rate": 0.00019956799318824776,
|
| 295 |
+
"loss": 0.3599,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.12678288431061807,
|
| 300 |
+
"eval_loss": 0.31717613339424133,
|
| 301 |
+
"eval_runtime": 878.4135,
|
| 302 |
+
"eval_samples_per_second": 4.554,
|
| 303 |
+
"eval_steps_per_second": 0.569,
|
| 304 |
+
"step": 200
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.12995245641838352,
|
| 308 |
+
"grad_norm": 0.32366959300654363,
|
| 309 |
+
"learning_rate": 0.00019945911021853818,
|
| 310 |
+
"loss": 0.2671,
|
| 311 |
+
"step": 205
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.13312202852614896,
|
| 315 |
+
"grad_norm": 0.34183927553766114,
|
| 316 |
+
"learning_rate": 0.00019933803973566102,
|
| 317 |
+
"loss": 0.3491,
|
| 318 |
+
"step": 210
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.13629160063391443,
|
| 322 |
+
"grad_norm": 0.355629049879592,
|
| 323 |
+
"learning_rate": 0.0001992047965753422,
|
| 324 |
+
"loss": 0.2778,
|
| 325 |
+
"step": 215
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.13946117274167988,
|
| 329 |
+
"grad_norm": 0.31194706241410036,
|
| 330 |
+
"learning_rate": 0.00019905939706492238,
|
| 331 |
+
"loss": 0.3278,
|
| 332 |
+
"step": 220
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.14263074484944532,
|
| 336 |
+
"grad_norm": 0.37190501088914274,
|
| 337 |
+
"learning_rate": 0.0001989018590213561,
|
| 338 |
+
"loss": 0.3757,
|
| 339 |
+
"step": 225
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.14580031695721077,
|
| 343 |
+
"grad_norm": 0.30859177154159206,
|
| 344 |
+
"learning_rate": 0.00019873220174902858,
|
| 345 |
+
"loss": 0.2952,
|
| 346 |
+
"step": 230
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.14896988906497624,
|
| 350 |
+
"grad_norm": 0.4072493051692793,
|
| 351 |
+
"learning_rate": 0.0001985504460373903,
|
| 352 |
+
"loss": 0.3576,
|
| 353 |
+
"step": 235
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.15213946117274169,
|
| 357 |
+
"grad_norm": 0.3117614582623609,
|
| 358 |
+
"learning_rate": 0.00019835661415840928,
|
| 359 |
+
"loss": 0.3127,
|
| 360 |
+
"step": 240
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.15530903328050713,
|
| 364 |
+
"grad_norm": 0.3433870206019631,
|
| 365 |
+
"learning_rate": 0.00019815072986384218,
|
| 366 |
+
"loss": 0.3424,
|
| 367 |
+
"step": 245
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.15847860538827258,
|
| 371 |
+
"grad_norm": 0.3252374107324197,
|
| 372 |
+
"learning_rate": 0.0001979328183823236,
|
| 373 |
+
"loss": 0.3509,
|
| 374 |
+
"step": 250
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.16164817749603805,
|
| 378 |
+
"grad_norm": 0.32574757253252834,
|
| 379 |
+
"learning_rate": 0.00019770290641627468,
|
| 380 |
+
"loss": 0.2913,
|
| 381 |
+
"step": 255
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.1648177496038035,
|
| 385 |
+
"grad_norm": 0.37343408069668577,
|
| 386 |
+
"learning_rate": 0.00019746102213863114,
|
| 387 |
+
"loss": 0.3524,
|
| 388 |
+
"step": 260
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.16798732171156894,
|
| 392 |
+
"grad_norm": 0.30197216412790706,
|
| 393 |
+
"learning_rate": 0.00019720719518939083,
|
| 394 |
+
"loss": 0.295,
|
| 395 |
+
"step": 265
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.17115689381933438,
|
| 399 |
+
"grad_norm": 0.37750434171669517,
|
| 400 |
+
"learning_rate": 0.00019694145667198195,
|
| 401 |
+
"loss": 0.3215,
|
| 402 |
+
"step": 270
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.17432646592709986,
|
| 406 |
+
"grad_norm": 0.3368196048030473,
|
| 407 |
+
"learning_rate": 0.0001966638391494514,
|
| 408 |
+
"loss": 0.35,
|
| 409 |
+
"step": 275
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.1774960380348653,
|
| 413 |
+
"grad_norm": 0.3232595651729065,
|
| 414 |
+
"learning_rate": 0.0001963743766404749,
|
| 415 |
+
"loss": 0.2637,
|
| 416 |
+
"step": 280
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.18066561014263074,
|
| 420 |
+
"grad_norm": 0.32199548202560035,
|
| 421 |
+
"learning_rate": 0.00019607310461518818,
|
| 422 |
+
"loss": 0.3262,
|
| 423 |
+
"step": 285
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.1838351822503962,
|
| 427 |
+
"grad_norm": 0.29117926540088634,
|
| 428 |
+
"learning_rate": 0.0001957600599908406,
|
| 429 |
+
"loss": 0.3129,
|
| 430 |
+
"step": 290
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.18700475435816163,
|
| 434 |
+
"grad_norm": 0.2836794081153409,
|
| 435 |
+
"learning_rate": 0.00019543528112727146,
|
| 436 |
+
"loss": 0.3207,
|
| 437 |
+
"step": 295
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.1901743264659271,
|
| 441 |
+
"grad_norm": 0.37478385305484463,
|
| 442 |
+
"learning_rate": 0.0001950988078222093,
|
| 443 |
+
"loss": 0.3503,
|
| 444 |
+
"step": 300
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.19334389857369255,
|
| 448 |
+
"grad_norm": 0.3323790483161259,
|
| 449 |
+
"learning_rate": 0.00019475068130639543,
|
| 450 |
+
"loss": 0.2873,
|
| 451 |
+
"step": 305
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.196513470681458,
|
| 455 |
+
"grad_norm": 0.31045326503955184,
|
| 456 |
+
"learning_rate": 0.0001943909442385313,
|
| 457 |
+
"loss": 0.3379,
|
| 458 |
+
"step": 310
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.19968304278922344,
|
| 462 |
+
"grad_norm": 0.295428110940092,
|
| 463 |
+
"learning_rate": 0.00019401964070005144,
|
| 464 |
+
"loss": 0.2913,
|
| 465 |
+
"step": 315
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.20285261489698891,
|
| 469 |
+
"grad_norm": 0.31381749704770145,
|
| 470 |
+
"learning_rate": 0.00019363681618972164,
|
| 471 |
+
"loss": 0.3167,
|
| 472 |
+
"step": 320
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.20602218700475436,
|
| 476 |
+
"grad_norm": 0.3799683908480184,
|
| 477 |
+
"learning_rate": 0.00019324251761806374,
|
| 478 |
+
"loss": 0.3203,
|
| 479 |
+
"step": 325
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.2091917591125198,
|
| 483 |
+
"grad_norm": 0.25669447806119594,
|
| 484 |
+
"learning_rate": 0.00019283679330160726,
|
| 485 |
+
"loss": 0.2598,
|
| 486 |
+
"step": 330
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.21236133122028525,
|
| 490 |
+
"grad_norm": 0.3253285501894849,
|
| 491 |
+
"learning_rate": 0.00019241969295696879,
|
| 492 |
+
"loss": 0.321,
|
| 493 |
+
"step": 335
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.21553090332805072,
|
| 497 |
+
"grad_norm": 0.3015776648780859,
|
| 498 |
+
"learning_rate": 0.0001919912676947598,
|
| 499 |
+
"loss": 0.2912,
|
| 500 |
+
"step": 340
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.21870047543581617,
|
| 504 |
+
"grad_norm": 0.3548152436637532,
|
| 505 |
+
"learning_rate": 0.00019155157001332374,
|
| 506 |
+
"loss": 0.3398,
|
| 507 |
+
"step": 345
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.2218700475435816,
|
| 511 |
+
"grad_norm": 0.3562179525646546,
|
| 512 |
+
"learning_rate": 0.00019110065379230289,
|
| 513 |
+
"loss": 0.3575,
|
| 514 |
+
"step": 350
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.22503961965134706,
|
| 518 |
+
"grad_norm": 0.33759944051182883,
|
| 519 |
+
"learning_rate": 0.00019063857428603615,
|
| 520 |
+
"loss": 0.2644,
|
| 521 |
+
"step": 355
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.22820919175911253,
|
| 525 |
+
"grad_norm": 0.3478332359179607,
|
| 526 |
+
"learning_rate": 0.00019016538811678823,
|
| 527 |
+
"loss": 0.3421,
|
| 528 |
+
"step": 360
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.23137876386687797,
|
| 532 |
+
"grad_norm": 0.3107602080624315,
|
| 533 |
+
"learning_rate": 0.0001896811532678113,
|
| 534 |
+
"loss": 0.262,
|
| 535 |
+
"step": 365
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.23454833597464342,
|
| 539 |
+
"grad_norm": 0.26971775917740104,
|
| 540 |
+
"learning_rate": 0.00018918592907623985,
|
| 541 |
+
"loss": 0.3378,
|
| 542 |
+
"step": 370
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.23771790808240886,
|
| 546 |
+
"grad_norm": 0.32413332448217697,
|
| 547 |
+
"learning_rate": 0.00018867977622581957,
|
| 548 |
+
"loss": 0.3316,
|
| 549 |
+
"step": 375
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.24088748019017434,
|
| 553 |
+
"grad_norm": 0.3522975093101741,
|
| 554 |
+
"learning_rate": 0.00018816275673947148,
|
| 555 |
+
"loss": 0.2678,
|
| 556 |
+
"step": 380
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.24405705229793978,
|
| 560 |
+
"grad_norm": 0.31661852350790726,
|
| 561 |
+
"learning_rate": 0.00018763493397169146,
|
| 562 |
+
"loss": 0.3275,
|
| 563 |
+
"step": 385
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.24722662440570523,
|
| 567 |
+
"grad_norm": 0.27090727261610936,
|
| 568 |
+
"learning_rate": 0.00018709637260078729,
|
| 569 |
+
"loss": 0.2858,
|
| 570 |
+
"step": 390
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.25039619651347067,
|
| 574 |
+
"grad_norm": 0.3143474617991223,
|
| 575 |
+
"learning_rate": 0.0001865471386209527,
|
| 576 |
+
"loss": 0.3317,
|
| 577 |
+
"step": 395
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.25356576862123614,
|
| 581 |
+
"grad_norm": 0.48811153855723693,
|
| 582 |
+
"learning_rate": 0.000185987299334181,
|
| 583 |
+
"loss": 0.3295,
|
| 584 |
+
"step": 400
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.25356576862123614,
|
| 588 |
+
"eval_loss": 0.29194891452789307,
|
| 589 |
+
"eval_runtime": 872.9978,
|
| 590 |
+
"eval_samples_per_second": 4.582,
|
| 591 |
+
"eval_steps_per_second": 0.573,
|
| 592 |
+
"step": 400
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.25673534072900156,
|
| 596 |
+
"grad_norm": 0.31755342222995686,
|
| 597 |
+
"learning_rate": 0.00018541692334201771,
|
| 598 |
+
"loss": 0.2643,
|
| 599 |
+
"step": 405
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 0.25990491283676703,
|
| 603 |
+
"grad_norm": 0.34778059073770806,
|
| 604 |
+
"learning_rate": 0.0001848360805371544,
|
| 605 |
+
"loss": 0.3339,
|
| 606 |
+
"step": 410
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 0.2630744849445325,
|
| 610 |
+
"grad_norm": 0.3183073063986642,
|
| 611 |
+
"learning_rate": 0.00018424484209486416,
|
| 612 |
+
"loss": 0.2673,
|
| 613 |
+
"step": 415
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 0.2662440570522979,
|
| 617 |
+
"grad_norm": 0.2788199901083398,
|
| 618 |
+
"learning_rate": 0.00018364328046428,
|
| 619 |
+
"loss": 0.3272,
|
| 620 |
+
"step": 420
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.2694136291600634,
|
| 624 |
+
"grad_norm": 0.3666143727147526,
|
| 625 |
+
"learning_rate": 0.00018303146935951689,
|
| 626 |
+
"loss": 0.3247,
|
| 627 |
+
"step": 425
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 0.27258320126782887,
|
| 631 |
+
"grad_norm": 0.28586548327038175,
|
| 632 |
+
"learning_rate": 0.00018240948375063926,
|
| 633 |
+
"loss": 0.2792,
|
| 634 |
+
"step": 430
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.2757527733755943,
|
| 638 |
+
"grad_norm": 0.9727255846044429,
|
| 639 |
+
"learning_rate": 0.00018177739985447412,
|
| 640 |
+
"loss": 0.3485,
|
| 641 |
+
"step": 435
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.27892234548335976,
|
| 645 |
+
"grad_norm": 0.29065854553956355,
|
| 646 |
+
"learning_rate": 0.0001811352951252717,
|
| 647 |
+
"loss": 0.2729,
|
| 648 |
+
"step": 440
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.2820919175911252,
|
| 652 |
+
"grad_norm": 0.320575993183303,
|
| 653 |
+
"learning_rate": 0.0001804832482452142,
|
| 654 |
+
"loss": 0.3354,
|
| 655 |
+
"step": 445
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.28526148969889065,
|
| 659 |
+
"grad_norm": 0.34869737354697955,
|
| 660 |
+
"learning_rate": 0.0001798213391147746,
|
| 661 |
+
"loss": 0.3385,
|
| 662 |
+
"step": 450
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.2884310618066561,
|
| 666 |
+
"grad_norm": 0.31478642211651564,
|
| 667 |
+
"learning_rate": 0.00017914964884292544,
|
| 668 |
+
"loss": 0.3133,
|
| 669 |
+
"step": 455
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.29160063391442154,
|
| 673 |
+
"grad_norm": 0.36834278711947965,
|
| 674 |
+
"learning_rate": 0.0001784682597372,
|
| 675 |
+
"loss": 0.3593,
|
| 676 |
+
"step": 460
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.294770206022187,
|
| 680 |
+
"grad_norm": 0.2791902388221146,
|
| 681 |
+
"learning_rate": 0.00017777725529360676,
|
| 682 |
+
"loss": 0.3005,
|
| 683 |
+
"step": 465
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.2979397781299525,
|
| 687 |
+
"grad_norm": 0.30096452678752406,
|
| 688 |
+
"learning_rate": 0.00017707672018639758,
|
| 689 |
+
"loss": 0.3354,
|
| 690 |
+
"step": 470
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.3011093502377179,
|
| 694 |
+
"grad_norm": 0.3708048891578612,
|
| 695 |
+
"learning_rate": 0.00017636674025769215,
|
| 696 |
+
"loss": 0.3147,
|
| 697 |
+
"step": 475
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.30427892234548337,
|
| 701 |
+
"grad_norm": 0.305209122691005,
|
| 702 |
+
"learning_rate": 0.00017564740250695904,
|
| 703 |
+
"loss": 0.2713,
|
| 704 |
+
"step": 480
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.3074484944532488,
|
| 708 |
+
"grad_norm": 0.3018873391630076,
|
| 709 |
+
"learning_rate": 0.0001749187950803549,
|
| 710 |
+
"loss": 0.3202,
|
| 711 |
+
"step": 485
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.31061806656101426,
|
| 715 |
+
"grad_norm": 0.3464422287874134,
|
| 716 |
+
"learning_rate": 0.00017418100725992316,
|
| 717 |
+
"loss": 0.3042,
|
| 718 |
+
"step": 490
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.31378763866877973,
|
| 722 |
+
"grad_norm": 0.31036543367721087,
|
| 723 |
+
"learning_rate": 0.00017343412945265382,
|
| 724 |
+
"loss": 0.3105,
|
| 725 |
+
"step": 495
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.31695721077654515,
|
| 729 |
+
"grad_norm": 0.3090116757558095,
|
| 730 |
+
"learning_rate": 0.00017267825317940493,
|
| 731 |
+
"loss": 0.3086,
|
| 732 |
+
"step": 500
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 0.3201267828843106,
|
| 736 |
+
"grad_norm": 0.32015559999952525,
|
| 737 |
+
"learning_rate": 0.00017191347106368797,
|
| 738 |
+
"loss": 0.2595,
|
| 739 |
+
"step": 505
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 0.3232963549920761,
|
| 743 |
+
"grad_norm": 0.28242640929152685,
|
| 744 |
+
"learning_rate": 0.0001711398768203178,
|
| 745 |
+
"loss": 0.3171,
|
| 746 |
+
"step": 510
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.3264659270998415,
|
| 750 |
+
"grad_norm": 0.3373697781712397,
|
| 751 |
+
"learning_rate": 0.00017035756524392924,
|
| 752 |
+
"loss": 0.2897,
|
| 753 |
+
"step": 515
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.329635499207607,
|
| 757 |
+
"grad_norm": 0.3187883343723006,
|
| 758 |
+
"learning_rate": 0.0001695666321973609,
|
| 759 |
+
"loss": 0.303,
|
| 760 |
+
"step": 520
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.3328050713153724,
|
| 764 |
+
"grad_norm": 0.4060972163443389,
|
| 765 |
+
"learning_rate": 0.00016876717459990862,
|
| 766 |
+
"loss": 0.3273,
|
| 767 |
+
"step": 525
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.3359746434231379,
|
| 771 |
+
"grad_norm": 0.2709960074426642,
|
| 772 |
+
"learning_rate": 0.0001679592904154489,
|
| 773 |
+
"loss": 0.2629,
|
| 774 |
+
"step": 530
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.33914421553090335,
|
| 778 |
+
"grad_norm": 0.2828719972128079,
|
| 779 |
+
"learning_rate": 0.00016714307864043487,
|
| 780 |
+
"loss": 0.2946,
|
| 781 |
+
"step": 535
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.34231378763866877,
|
| 785 |
+
"grad_norm": 0.29485357171410065,
|
| 786 |
+
"learning_rate": 0.00016631863929176524,
|
| 787 |
+
"loss": 0.2704,
|
| 788 |
+
"step": 540
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 0.34548335974643424,
|
| 792 |
+
"grad_norm": 0.3140677978027709,
|
| 793 |
+
"learning_rate": 0.00016548607339452853,
|
| 794 |
+
"loss": 0.3211,
|
| 795 |
+
"step": 545
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 0.3486529318541997,
|
| 799 |
+
"grad_norm": 0.30224374704766904,
|
| 800 |
+
"learning_rate": 0.00016464548296962373,
|
| 801 |
+
"loss": 0.3289,
|
| 802 |
+
"step": 550
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 0.3518225039619651,
|
| 806 |
+
"grad_norm": 0.3015178734291492,
|
| 807 |
+
"learning_rate": 0.0001637969710212588,
|
| 808 |
+
"loss": 0.262,
|
| 809 |
+
"step": 555
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.3549920760697306,
|
| 813 |
+
"grad_norm": 0.3261808476280464,
|
| 814 |
+
"learning_rate": 0.00016294064152432879,
|
| 815 |
+
"loss": 0.3524,
|
| 816 |
+
"step": 560
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 0.358161648177496,
|
| 820 |
+
"grad_norm": 0.30420040263110554,
|
| 821 |
+
"learning_rate": 0.00016207659941167485,
|
| 822 |
+
"loss": 0.2888,
|
| 823 |
+
"step": 565
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 0.3613312202852615,
|
| 827 |
+
"grad_norm": 0.29855740633395794,
|
| 828 |
+
"learning_rate": 0.00016120495056122622,
|
| 829 |
+
"loss": 0.3075,
|
| 830 |
+
"step": 570
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.36450079239302696,
|
| 834 |
+
"grad_norm": 0.3775755682614953,
|
| 835 |
+
"learning_rate": 0.00016032580178302583,
|
| 836 |
+
"loss": 0.3452,
|
| 837 |
+
"step": 575
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 0.3676703645007924,
|
| 841 |
+
"grad_norm": 0.3189277602131783,
|
| 842 |
+
"learning_rate": 0.00015943926080614235,
|
| 843 |
+
"loss": 0.2643,
|
| 844 |
+
"step": 580
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 0.37083993660855785,
|
| 848 |
+
"grad_norm": 0.32115548282274786,
|
| 849 |
+
"learning_rate": 0.00015854543626546915,
|
| 850 |
+
"loss": 0.3126,
|
| 851 |
+
"step": 585
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 0.37400950871632327,
|
| 855 |
+
"grad_norm": 0.29230296850863174,
|
| 856 |
+
"learning_rate": 0.00015764443768841234,
|
| 857 |
+
"loss": 0.2949,
|
| 858 |
+
"step": 590
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 0.37717908082408874,
|
| 862 |
+
"grad_norm": 0.32187057297721217,
|
| 863 |
+
"learning_rate": 0.0001567363754814696,
|
| 864 |
+
"loss": 0.3166,
|
| 865 |
+
"step": 595
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.3803486529318542,
|
| 869 |
+
"grad_norm": 0.3766752931165212,
|
| 870 |
+
"learning_rate": 0.0001558213609167012,
|
| 871 |
+
"loss": 0.323,
|
| 872 |
+
"step": 600
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.3803486529318542,
|
| 876 |
+
"eval_loss": 0.2788923680782318,
|
| 877 |
+
"eval_runtime": 873.5171,
|
| 878 |
+
"eval_samples_per_second": 4.579,
|
| 879 |
+
"eval_steps_per_second": 0.572,
|
| 880 |
+
"step": 600
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 0.38351822503961963,
|
| 884 |
+
"grad_norm": 0.31877960462977273,
|
| 885 |
+
"learning_rate": 0.00015489950611809484,
|
| 886 |
+
"loss": 0.2803,
|
| 887 |
+
"step": 605
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 0.3866877971473851,
|
| 891 |
+
"grad_norm": 0.2903622851026156,
|
| 892 |
+
"learning_rate": 0.00015397092404782642,
|
| 893 |
+
"loss": 0.3178,
|
| 894 |
+
"step": 610
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.3898573692551506,
|
| 898 |
+
"grad_norm": 0.2639727101749139,
|
| 899 |
+
"learning_rate": 0.00015303572849241764,
|
| 900 |
+
"loss": 0.2703,
|
| 901 |
+
"step": 615
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 0.393026941362916,
|
| 905 |
+
"grad_norm": 0.3491709894849581,
|
| 906 |
+
"learning_rate": 0.00015209403404879303,
|
| 907 |
+
"loss": 0.3049,
|
| 908 |
+
"step": 620
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 0.39619651347068147,
|
| 912 |
+
"grad_norm": 0.3651420024997032,
|
| 913 |
+
"learning_rate": 0.00015114595611023744,
|
| 914 |
+
"loss": 0.3265,
|
| 915 |
+
"step": 625
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 0.3993660855784469,
|
| 919 |
+
"grad_norm": 0.3071330073578763,
|
| 920 |
+
"learning_rate": 0.0001501916108522558,
|
| 921 |
+
"loss": 0.2645,
|
| 922 |
+
"step": 630
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 0.40253565768621236,
|
| 926 |
+
"grad_norm": 0.2739471545543727,
|
| 927 |
+
"learning_rate": 0.00014923111521833758,
|
| 928 |
+
"loss": 0.3035,
|
| 929 |
+
"step": 635
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.40570522979397783,
|
| 933 |
+
"grad_norm": 0.30630113259525843,
|
| 934 |
+
"learning_rate": 0.00014826458690562642,
|
| 935 |
+
"loss": 0.2606,
|
| 936 |
+
"step": 640
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 0.40887480190174325,
|
| 940 |
+
"grad_norm": 0.2988843883769528,
|
| 941 |
+
"learning_rate": 0.00014729214435049793,
|
| 942 |
+
"loss": 0.3111,
|
| 943 |
+
"step": 645
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 0.4120443740095087,
|
| 947 |
+
"grad_norm": 0.3110979862585215,
|
| 948 |
+
"learning_rate": 0.0001463139067140468,
|
| 949 |
+
"loss": 0.2948,
|
| 950 |
+
"step": 650
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 0.4152139461172742,
|
| 954 |
+
"grad_norm": 0.30767657253531316,
|
| 955 |
+
"learning_rate": 0.0001453299938674849,
|
| 956 |
+
"loss": 0.2638,
|
| 957 |
+
"step": 655
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 0.4183835182250396,
|
| 961 |
+
"grad_norm": 0.27014842841388653,
|
| 962 |
+
"learning_rate": 0.00014434052637745257,
|
| 963 |
+
"loss": 0.2819,
|
| 964 |
+
"step": 660
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 0.4215530903328051,
|
| 968 |
+
"grad_norm": 0.2739393681355767,
|
| 969 |
+
"learning_rate": 0.00014334562549124467,
|
| 970 |
+
"loss": 0.2466,
|
| 971 |
+
"step": 665
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 0.4247226624405705,
|
| 975 |
+
"grad_norm": 0.31758998023523244,
|
| 976 |
+
"learning_rate": 0.00014234541312195323,
|
| 977 |
+
"loss": 0.2873,
|
| 978 |
+
"step": 670
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 0.42789223454833597,
|
| 982 |
+
"grad_norm": 0.39847849128188423,
|
| 983 |
+
"learning_rate": 0.00014134001183352832,
|
| 984 |
+
"loss": 0.2979,
|
| 985 |
+
"step": 675
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 0.43106180665610144,
|
| 989 |
+
"grad_norm": 0.30950118355401873,
|
| 990 |
+
"learning_rate": 0.00014032954482575937,
|
| 991 |
+
"loss": 0.2617,
|
| 992 |
+
"step": 680
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 0.43423137876386686,
|
| 996 |
+
"grad_norm": 0.3260587574739946,
|
| 997 |
+
"learning_rate": 0.0001393141359191787,
|
| 998 |
+
"loss": 0.3109,
|
| 999 |
+
"step": 685
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 0.43740095087163233,
|
| 1003 |
+
"grad_norm": 0.3114375419997854,
|
| 1004 |
+
"learning_rate": 0.00013829390953988853,
|
| 1005 |
+
"loss": 0.2845,
|
| 1006 |
+
"step": 690
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 0.4405705229793978,
|
| 1010 |
+
"grad_norm": 0.30019871836883555,
|
| 1011 |
+
"learning_rate": 0.00013726899070431423,
|
| 1012 |
+
"loss": 0.324,
|
| 1013 |
+
"step": 695
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 0.4437400950871632,
|
| 1017 |
+
"grad_norm": 0.38021042516470643,
|
| 1018 |
+
"learning_rate": 0.00013623950500388506,
|
| 1019 |
+
"loss": 0.3269,
|
| 1020 |
+
"step": 700
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 0.4469096671949287,
|
| 1024 |
+
"grad_norm": 0.3089060241706131,
|
| 1025 |
+
"learning_rate": 0.00013520557858964446,
|
| 1026 |
+
"loss": 0.2584,
|
| 1027 |
+
"step": 705
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 0.4500792393026941,
|
| 1031 |
+
"grad_norm": 0.27984586622582663,
|
| 1032 |
+
"learning_rate": 0.00013416733815679166,
|
| 1033 |
+
"loss": 0.2909,
|
| 1034 |
+
"step": 710
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 0.4532488114104596,
|
| 1038 |
+
"grad_norm": 0.2923559292409706,
|
| 1039 |
+
"learning_rate": 0.00013312491092915682,
|
| 1040 |
+
"loss": 0.2489,
|
| 1041 |
+
"step": 715
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 0.45641838351822506,
|
| 1045 |
+
"grad_norm": 0.29223045315786345,
|
| 1046 |
+
"learning_rate": 0.00013207842464361125,
|
| 1047 |
+
"loss": 0.3135,
|
| 1048 |
+
"step": 720
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 0.4595879556259905,
|
| 1052 |
+
"grad_norm": 0.33907899924090856,
|
| 1053 |
+
"learning_rate": 0.00013102800753441487,
|
| 1054 |
+
"loss": 0.3148,
|
| 1055 |
+
"step": 725
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 0.46275752773375595,
|
| 1059 |
+
"grad_norm": 0.26110455456342696,
|
| 1060 |
+
"learning_rate": 0.00012997378831750242,
|
| 1061 |
+
"loss": 0.2505,
|
| 1062 |
+
"step": 730
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 0.4659270998415214,
|
| 1066 |
+
"grad_norm": 0.2855563878095534,
|
| 1067 |
+
"learning_rate": 0.00012891589617471122,
|
| 1068 |
+
"loss": 0.322,
|
| 1069 |
+
"step": 735
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 0.46909667194928684,
|
| 1073 |
+
"grad_norm": 0.27089962197787903,
|
| 1074 |
+
"learning_rate": 0.00012785446073795118,
|
| 1075 |
+
"loss": 0.2629,
|
| 1076 |
+
"step": 740
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 0.4722662440570523,
|
| 1080 |
+
"grad_norm": 0.2787588891548799,
|
| 1081 |
+
"learning_rate": 0.00012678961207332015,
|
| 1082 |
+
"loss": 0.3071,
|
| 1083 |
+
"step": 745
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 0.4754358161648177,
|
| 1087 |
+
"grad_norm": 0.35249049637057156,
|
| 1088 |
+
"learning_rate": 0.00012572148066516584,
|
| 1089 |
+
"loss": 0.3265,
|
| 1090 |
+
"step": 750
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 0.4786053882725832,
|
| 1094 |
+
"grad_norm": 0.33307560406452336,
|
| 1095 |
+
"learning_rate": 0.00012465019740009662,
|
| 1096 |
+
"loss": 0.2403,
|
| 1097 |
+
"step": 755
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 0.48177496038034867,
|
| 1101 |
+
"grad_norm": 0.3035753509057755,
|
| 1102 |
+
"learning_rate": 0.00012357589355094275,
|
| 1103 |
+
"loss": 0.3057,
|
| 1104 |
+
"step": 760
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 0.4849445324881141,
|
| 1108 |
+
"grad_norm": 0.2950972689886197,
|
| 1109 |
+
"learning_rate": 0.00012249870076067067,
|
| 1110 |
+
"loss": 0.2637,
|
| 1111 |
+
"step": 765
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 0.48811410459587956,
|
| 1115 |
+
"grad_norm": 0.2713040409786771,
|
| 1116 |
+
"learning_rate": 0.00012141875102625167,
|
| 1117 |
+
"loss": 0.3196,
|
| 1118 |
+
"step": 770
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 0.49128367670364503,
|
| 1122 |
+
"grad_norm": 0.37005187803966516,
|
| 1123 |
+
"learning_rate": 0.00012033617668248723,
|
| 1124 |
+
"loss": 0.3265,
|
| 1125 |
+
"step": 775
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 0.49445324881141045,
|
| 1129 |
+
"grad_norm": 0.3678796577106568,
|
| 1130 |
+
"learning_rate": 0.00011925111038579309,
|
| 1131 |
+
"loss": 0.2283,
|
| 1132 |
+
"step": 780
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 0.4976228209191759,
|
| 1136 |
+
"grad_norm": 0.3021844529595635,
|
| 1137 |
+
"learning_rate": 0.00011816368509794364,
|
| 1138 |
+
"loss": 0.2967,
|
| 1139 |
+
"step": 785
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 0.5007923930269413,
|
| 1143 |
+
"grad_norm": 0.3028161473676034,
|
| 1144 |
+
"learning_rate": 0.00011707403406977928,
|
| 1145 |
+
"loss": 0.2841,
|
| 1146 |
+
"step": 790
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 0.5039619651347068,
|
| 1150 |
+
"grad_norm": 0.27418964538735746,
|
| 1151 |
+
"learning_rate": 0.00011598229082487784,
|
| 1152 |
+
"loss": 0.2803,
|
| 1153 |
+
"step": 795
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 0.5071315372424723,
|
| 1157 |
+
"grad_norm": 0.3426638434156249,
|
| 1158 |
+
"learning_rate": 0.0001148885891431932,
|
| 1159 |
+
"loss": 0.3274,
|
| 1160 |
+
"step": 800
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 0.5071315372424723,
|
| 1164 |
+
"eval_loss": 0.26855266094207764,
|
| 1165 |
+
"eval_runtime": 873.628,
|
| 1166 |
+
"eval_samples_per_second": 4.579,
|
| 1167 |
+
"eval_steps_per_second": 0.572,
|
| 1168 |
+
"step": 800
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 0.5103011093502378,
|
| 1172 |
+
"grad_norm": 0.2681269338020656,
|
| 1173 |
+
"learning_rate": 0.00011379306304466198,
|
| 1174 |
+
"loss": 0.2381,
|
| 1175 |
+
"step": 805
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 0.5134706814580031,
|
| 1179 |
+
"grad_norm": 0.2987060218422062,
|
| 1180 |
+
"learning_rate": 0.00011269584677278102,
|
| 1181 |
+
"loss": 0.3076,
|
| 1182 |
+
"step": 810
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 0.5166402535657686,
|
| 1186 |
+
"grad_norm": 0.2804222341073312,
|
| 1187 |
+
"learning_rate": 0.00011159707477815755,
|
| 1188 |
+
"loss": 0.2395,
|
| 1189 |
+
"step": 815
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 0.5198098256735341,
|
| 1193 |
+
"grad_norm": 0.25835895356413513,
|
| 1194 |
+
"learning_rate": 0.00011049688170203383,
|
| 1195 |
+
"loss": 0.3041,
|
| 1196 |
+
"step": 820
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 0.5229793977812995,
|
| 1200 |
+
"grad_norm": 0.3313190058494361,
|
| 1201 |
+
"learning_rate": 0.00010939540235978845,
|
| 1202 |
+
"loss": 0.297,
|
| 1203 |
+
"step": 825
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 0.526148969889065,
|
| 1207 |
+
"grad_norm": 0.2564972143294916,
|
| 1208 |
+
"learning_rate": 0.00010829277172441648,
|
| 1209 |
+
"loss": 0.2359,
|
| 1210 |
+
"step": 830
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 0.5293185419968305,
|
| 1214 |
+
"grad_norm": 0.31632766018739716,
|
| 1215 |
+
"learning_rate": 0.00010718912490998991,
|
| 1216 |
+
"loss": 0.3112,
|
| 1217 |
+
"step": 835
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 0.5324881141045958,
|
| 1221 |
+
"grad_norm": 0.2738970193614327,
|
| 1222 |
+
"learning_rate": 0.00010608459715510139,
|
| 1223 |
+
"loss": 0.2416,
|
| 1224 |
+
"step": 840
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 0.5356576862123613,
|
| 1228 |
+
"grad_norm": 0.35306801364530893,
|
| 1229 |
+
"learning_rate": 0.00010497932380629207,
|
| 1230 |
+
"loss": 0.3334,
|
| 1231 |
+
"step": 845
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 0.5388272583201268,
|
| 1235 |
+
"grad_norm": 0.3617753781992424,
|
| 1236 |
+
"learning_rate": 0.00010387344030146665,
|
| 1237 |
+
"loss": 0.3071,
|
| 1238 |
+
"step": 850
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 0.5419968304278923,
|
| 1242 |
+
"grad_norm": 0.284695185318866,
|
| 1243 |
+
"learning_rate": 0.0001027670821532971,
|
| 1244 |
+
"loss": 0.2516,
|
| 1245 |
+
"step": 855
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 0.5451664025356577,
|
| 1249 |
+
"grad_norm": 0.28641499966999695,
|
| 1250 |
+
"learning_rate": 0.00010166038493261722,
|
| 1251 |
+
"loss": 0.3268,
|
| 1252 |
+
"step": 860
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 0.5483359746434231,
|
| 1256 |
+
"grad_norm": 0.29940254299061986,
|
| 1257 |
+
"learning_rate": 0.00010055348425181,
|
| 1258 |
+
"loss": 0.2667,
|
| 1259 |
+
"step": 865
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 0.5515055467511886,
|
| 1263 |
+
"grad_norm": 0.33784906825030664,
|
| 1264 |
+
"learning_rate": 9.944651574819003e-05,
|
| 1265 |
+
"loss": 0.3006,
|
| 1266 |
+
"step": 870
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 0.554675118858954,
|
| 1270 |
+
"grad_norm": 0.33800198210916443,
|
| 1271 |
+
"learning_rate": 9.83396150673828e-05,
|
| 1272 |
+
"loss": 0.3009,
|
| 1273 |
+
"step": 875
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 0.5578446909667195,
|
| 1277 |
+
"grad_norm": 0.27814752259908526,
|
| 1278 |
+
"learning_rate": 9.72329178467029e-05,
|
| 1279 |
+
"loss": 0.25,
|
| 1280 |
+
"step": 880
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 0.561014263074485,
|
| 1284 |
+
"grad_norm": 0.3120985607406773,
|
| 1285 |
+
"learning_rate": 9.612655969853336e-05,
|
| 1286 |
+
"loss": 0.3079,
|
| 1287 |
+
"step": 885
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 0.5641838351822503,
|
| 1291 |
+
"grad_norm": 0.32270045792226343,
|
| 1292 |
+
"learning_rate": 9.502067619370794e-05,
|
| 1293 |
+
"loss": 0.2465,
|
| 1294 |
+
"step": 890
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 0.5673534072900158,
|
| 1298 |
+
"grad_norm": 0.2522429392869884,
|
| 1299 |
+
"learning_rate": 9.391540284489862e-05,
|
| 1300 |
+
"loss": 0.3049,
|
| 1301 |
+
"step": 895
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 0.5705229793977813,
|
| 1305 |
+
"grad_norm": 0.32479021947356745,
|
| 1306 |
+
"learning_rate": 9.281087509001011e-05,
|
| 1307 |
+
"loss": 0.3109,
|
| 1308 |
+
"step": 900
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 0.5736925515055468,
|
| 1312 |
+
"grad_norm": 0.3071871099500722,
|
| 1313 |
+
"learning_rate": 9.170722827558358e-05,
|
| 1314 |
+
"loss": 0.2566,
|
| 1315 |
+
"step": 905
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 0.5768621236133122,
|
| 1319 |
+
"grad_norm": 0.2808358292017096,
|
| 1320 |
+
"learning_rate": 9.060459764021156e-05,
|
| 1321 |
+
"loss": 0.2981,
|
| 1322 |
+
"step": 910
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 0.5800316957210776,
|
| 1326 |
+
"grad_norm": 0.36613518181258947,
|
| 1327 |
+
"learning_rate": 8.950311829796619e-05,
|
| 1328 |
+
"loss": 0.2812,
|
| 1329 |
+
"step": 915
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 0.5832012678288431,
|
| 1333 |
+
"grad_norm": 0.29120302112196544,
|
| 1334 |
+
"learning_rate": 8.840292522184247e-05,
|
| 1335 |
+
"loss": 0.2958,
|
| 1336 |
+
"step": 920
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 0.5863708399366085,
|
| 1340 |
+
"grad_norm": 0.3008146054202439,
|
| 1341 |
+
"learning_rate": 8.730415322721897e-05,
|
| 1342 |
+
"loss": 0.3119,
|
| 1343 |
+
"step": 925
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 0.589540412044374,
|
| 1347 |
+
"grad_norm": 0.30809505125548203,
|
| 1348 |
+
"learning_rate": 8.620693695533803e-05,
|
| 1349 |
+
"loss": 0.2603,
|
| 1350 |
+
"step": 930
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 0.5927099841521395,
|
| 1354 |
+
"grad_norm": 0.3464042931932695,
|
| 1355 |
+
"learning_rate": 8.511141085680683e-05,
|
| 1356 |
+
"loss": 0.3217,
|
| 1357 |
+
"step": 935
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"epoch": 0.595879556259905,
|
| 1361 |
+
"grad_norm": 0.28395404105986655,
|
| 1362 |
+
"learning_rate": 8.401770917512221e-05,
|
| 1363 |
+
"loss": 0.2339,
|
| 1364 |
+
"step": 940
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 0.5990491283676703,
|
| 1368 |
+
"grad_norm": 0.32456815689823176,
|
| 1369 |
+
"learning_rate": 8.292596593022075e-05,
|
| 1370 |
+
"loss": 0.2761,
|
| 1371 |
+
"step": 945
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 0.6022187004754358,
|
| 1375 |
+
"grad_norm": 0.35814205267620147,
|
| 1376 |
+
"learning_rate": 8.183631490205637e-05,
|
| 1377 |
+
"loss": 0.3064,
|
| 1378 |
+
"step": 950
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 0.6053882725832013,
|
| 1382 |
+
"grad_norm": 0.3307025804465351,
|
| 1383 |
+
"learning_rate": 8.074888961420695e-05,
|
| 1384 |
+
"loss": 0.2317,
|
| 1385 |
+
"step": 955
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 0.6085578446909667,
|
| 1389 |
+
"grad_norm": 0.3035093202164917,
|
| 1390 |
+
"learning_rate": 7.966382331751277e-05,
|
| 1391 |
+
"loss": 0.3024,
|
| 1392 |
+
"step": 960
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 0.6117274167987322,
|
| 1396 |
+
"grad_norm": 0.23483953416505404,
|
| 1397 |
+
"learning_rate": 7.858124897374837e-05,
|
| 1398 |
+
"loss": 0.2616,
|
| 1399 |
+
"step": 965
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 0.6148969889064976,
|
| 1403 |
+
"grad_norm": 0.24795445024402282,
|
| 1404 |
+
"learning_rate": 7.750129923932939e-05,
|
| 1405 |
+
"loss": 0.2889,
|
| 1406 |
+
"step": 970
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 0.618066561014263,
|
| 1410 |
+
"grad_norm": 0.39470726118892546,
|
| 1411 |
+
"learning_rate": 7.642410644905726e-05,
|
| 1412 |
+
"loss": 0.3255,
|
| 1413 |
+
"step": 975
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 0.6212361331220285,
|
| 1417 |
+
"grad_norm": 0.28578857562483734,
|
| 1418 |
+
"learning_rate": 7.534980259990341e-05,
|
| 1419 |
+
"loss": 0.2177,
|
| 1420 |
+
"step": 980
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 0.624405705229794,
|
| 1424 |
+
"grad_norm": 0.293120691065387,
|
| 1425 |
+
"learning_rate": 7.427851933483418e-05,
|
| 1426 |
+
"loss": 0.3008,
|
| 1427 |
+
"step": 985
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 0.6275752773375595,
|
| 1431 |
+
"grad_norm": 0.28050824031198807,
|
| 1432 |
+
"learning_rate": 7.321038792667987e-05,
|
| 1433 |
+
"loss": 0.2617,
|
| 1434 |
+
"step": 990
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 0.6307448494453248,
|
| 1438 |
+
"grad_norm": 0.3421819179459905,
|
| 1439 |
+
"learning_rate": 7.214553926204883e-05,
|
| 1440 |
+
"loss": 0.2827,
|
| 1441 |
+
"step": 995
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 0.6339144215530903,
|
| 1445 |
+
"grad_norm": 0.3825000717076991,
|
| 1446 |
+
"learning_rate": 7.108410382528879e-05,
|
| 1447 |
+
"loss": 0.3171,
|
| 1448 |
+
"step": 1000
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 0.6339144215530903,
|
| 1452 |
+
"eval_loss": 0.2597305178642273,
|
| 1453 |
+
"eval_runtime": 873.3574,
|
| 1454 |
+
"eval_samples_per_second": 4.58,
|
| 1455 |
+
"eval_steps_per_second": 0.573,
|
| 1456 |
+
"step": 1000
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 0.6370839936608558,
|
| 1460 |
+
"grad_norm": 0.293460396656183,
|
| 1461 |
+
"learning_rate": 7.002621168249759e-05,
|
| 1462 |
+
"loss": 0.2297,
|
| 1463 |
+
"step": 1005
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 0.6402535657686212,
|
| 1467 |
+
"grad_norm": 0.3006160040194,
|
| 1468 |
+
"learning_rate": 6.897199246558514e-05,
|
| 1469 |
+
"loss": 0.2956,
|
| 1470 |
+
"step": 1010
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 0.6434231378763867,
|
| 1474 |
+
"grad_norm": 0.2791223126874652,
|
| 1475 |
+
"learning_rate": 6.792157535638874e-05,
|
| 1476 |
+
"loss": 0.2496,
|
| 1477 |
+
"step": 1015
|
| 1478 |
+
},
|
| 1479 |
+
{
|
| 1480 |
+
"epoch": 0.6465927099841522,
|
| 1481 |
+
"grad_norm": 0.2894662197144813,
|
| 1482 |
+
"learning_rate": 6.687508907084319e-05,
|
| 1483 |
+
"loss": 0.2866,
|
| 1484 |
+
"step": 1020
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 0.6497622820919176,
|
| 1488 |
+
"grad_norm": 0.33156274133370534,
|
| 1489 |
+
"learning_rate": 6.583266184320836e-05,
|
| 1490 |
+
"loss": 0.32,
|
| 1491 |
+
"step": 1025
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 0.652931854199683,
|
| 1495 |
+
"grad_norm": 0.3447301699746775,
|
| 1496 |
+
"learning_rate": 6.479442141035556e-05,
|
| 1497 |
+
"loss": 0.2555,
|
| 1498 |
+
"step": 1030
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 0.6561014263074485,
|
| 1502 |
+
"grad_norm": 0.3019937172628048,
|
| 1503 |
+
"learning_rate": 6.376049499611496e-05,
|
| 1504 |
+
"loss": 0.2632,
|
| 1505 |
+
"step": 1035
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 0.659270998415214,
|
| 1509 |
+
"grad_norm": 0.25047087286035274,
|
| 1510 |
+
"learning_rate": 6.273100929568578e-05,
|
| 1511 |
+
"loss": 0.2472,
|
| 1512 |
+
"step": 1040
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 0.6624405705229794,
|
| 1516 |
+
"grad_norm": 0.31801398649186896,
|
| 1517 |
+
"learning_rate": 6.170609046011151e-05,
|
| 1518 |
+
"loss": 0.2793,
|
| 1519 |
+
"step": 1045
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 0.6656101426307448,
|
| 1523 |
+
"grad_norm": 0.3464523898432614,
|
| 1524 |
+
"learning_rate": 6.068586408082133e-05,
|
| 1525 |
+
"loss": 0.3138,
|
| 1526 |
+
"step": 1050
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 0.6687797147385103,
|
| 1530 |
+
"grad_norm": 0.2919062799416737,
|
| 1531 |
+
"learning_rate": 5.9670455174240614e-05,
|
| 1532 |
+
"loss": 0.2427,
|
| 1533 |
+
"step": 1055
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 0.6719492868462758,
|
| 1537 |
+
"grad_norm": 0.29267872629520425,
|
| 1538 |
+
"learning_rate": 5.865998816647171e-05,
|
| 1539 |
+
"loss": 0.3038,
|
| 1540 |
+
"step": 1060
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 0.6751188589540412,
|
| 1544 |
+
"grad_norm": 0.27361822239828004,
|
| 1545 |
+
"learning_rate": 5.765458687804679e-05,
|
| 1546 |
+
"loss": 0.2566,
|
| 1547 |
+
"step": 1065
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 0.6782884310618067,
|
| 1551 |
+
"grad_norm": 0.3050132066017946,
|
| 1552 |
+
"learning_rate": 5.665437450875534e-05,
|
| 1553 |
+
"loss": 0.2752,
|
| 1554 |
+
"step": 1070
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 0.6814580031695721,
|
| 1558 |
+
"grad_norm": 0.3580338711915158,
|
| 1559 |
+
"learning_rate": 5.565947362254746e-05,
|
| 1560 |
+
"loss": 0.3331,
|
| 1561 |
+
"step": 1075
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 0.6846275752773375,
|
| 1565 |
+
"grad_norm": 0.26747930377415474,
|
| 1566 |
+
"learning_rate": 5.467000613251516e-05,
|
| 1567 |
+
"loss": 0.2429,
|
| 1568 |
+
"step": 1080
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 0.687797147385103,
|
| 1572 |
+
"grad_norm": 0.32226567868782413,
|
| 1573 |
+
"learning_rate": 5.368609328595323e-05,
|
| 1574 |
+
"loss": 0.3208,
|
| 1575 |
+
"step": 1085
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 0.6909667194928685,
|
| 1579 |
+
"grad_norm": 0.27314417996148593,
|
| 1580 |
+
"learning_rate": 5.270785564950208e-05,
|
| 1581 |
+
"loss": 0.2351,
|
| 1582 |
+
"step": 1090
|
| 1583 |
+
},
|
| 1584 |
+
{
|
| 1585 |
+
"epoch": 0.694136291600634,
|
| 1586 |
+
"grad_norm": 0.31179553442460595,
|
| 1587 |
+
"learning_rate": 5.1735413094373594e-05,
|
| 1588 |
+
"loss": 0.2791,
|
| 1589 |
+
"step": 1095
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 0.6973058637083994,
|
| 1593 |
+
"grad_norm": 0.2983027582550753,
|
| 1594 |
+
"learning_rate": 5.0768884781662465e-05,
|
| 1595 |
+
"loss": 0.3123,
|
| 1596 |
+
"step": 1100
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 0.7004754358161648,
|
| 1600 |
+
"grad_norm": 0.268619063810808,
|
| 1601 |
+
"learning_rate": 4.9808389147744195e-05,
|
| 1602 |
+
"loss": 0.2675,
|
| 1603 |
+
"step": 1105
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 0.7036450079239303,
|
| 1607 |
+
"grad_norm": 0.34151620569667657,
|
| 1608 |
+
"learning_rate": 4.885404388976261e-05,
|
| 1609 |
+
"loss": 0.3171,
|
| 1610 |
+
"step": 1110
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 0.7068145800316957,
|
| 1614 |
+
"grad_norm": 0.25963093128586956,
|
| 1615 |
+
"learning_rate": 4.790596595120699e-05,
|
| 1616 |
+
"loss": 0.2533,
|
| 1617 |
+
"step": 1115
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 0.7099841521394612,
|
| 1621 |
+
"grad_norm": 0.3373621924020373,
|
| 1622 |
+
"learning_rate": 4.696427150758238e-05,
|
| 1623 |
+
"loss": 0.3017,
|
| 1624 |
+
"step": 1120
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 0.7131537242472267,
|
| 1628 |
+
"grad_norm": 0.32633352666577314,
|
| 1629 |
+
"learning_rate": 4.6029075952173596e-05,
|
| 1630 |
+
"loss": 0.3052,
|
| 1631 |
+
"step": 1125
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 0.716323296354992,
|
| 1635 |
+
"grad_norm": 0.24971258370165642,
|
| 1636 |
+
"learning_rate": 4.510049388190518e-05,
|
| 1637 |
+
"loss": 0.2044,
|
| 1638 |
+
"step": 1130
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 0.7194928684627575,
|
| 1642 |
+
"grad_norm": 0.29602844393415106,
|
| 1643 |
+
"learning_rate": 4.417863908329884e-05,
|
| 1644 |
+
"loss": 0.2959,
|
| 1645 |
+
"step": 1135
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 0.722662440570523,
|
| 1649 |
+
"grad_norm": 0.23146594836780063,
|
| 1650 |
+
"learning_rate": 4.32636245185304e-05,
|
| 1651 |
+
"loss": 0.2252,
|
| 1652 |
+
"step": 1140
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 0.7258320126782885,
|
| 1656 |
+
"grad_norm": 0.2744736835188008,
|
| 1657 |
+
"learning_rate": 4.235556231158765e-05,
|
| 1658 |
+
"loss": 0.2884,
|
| 1659 |
+
"step": 1145
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 0.7290015847860539,
|
| 1663 |
+
"grad_norm": 0.27538990975844047,
|
| 1664 |
+
"learning_rate": 4.145456373453087e-05,
|
| 1665 |
+
"loss": 0.2981,
|
| 1666 |
+
"step": 1150
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 0.7321711568938193,
|
| 1670 |
+
"grad_norm": 0.3032208366026702,
|
| 1671 |
+
"learning_rate": 4.0560739193857625e-05,
|
| 1672 |
+
"loss": 0.2158,
|
| 1673 |
+
"step": 1155
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 0.7353407290015848,
|
| 1677 |
+
"grad_norm": 0.27204457210068295,
|
| 1678 |
+
"learning_rate": 3.96741982169742e-05,
|
| 1679 |
+
"loss": 0.3028,
|
| 1680 |
+
"step": 1160
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 0.7385103011093502,
|
| 1684 |
+
"grad_norm": 0.28301662262727184,
|
| 1685 |
+
"learning_rate": 3.8795049438773825e-05,
|
| 1686 |
+
"loss": 0.2946,
|
| 1687 |
+
"step": 1165
|
| 1688 |
+
},
|
| 1689 |
+
{
|
| 1690 |
+
"epoch": 0.7416798732171157,
|
| 1691 |
+
"grad_norm": 0.2884264535746388,
|
| 1692 |
+
"learning_rate": 3.7923400588325155e-05,
|
| 1693 |
+
"loss": 0.3015,
|
| 1694 |
+
"step": 1170
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 0.7448494453248812,
|
| 1698 |
+
"grad_norm": 0.3186549926460967,
|
| 1699 |
+
"learning_rate": 3.7059358475671224e-05,
|
| 1700 |
+
"loss": 0.2773,
|
| 1701 |
+
"step": 1175
|
| 1702 |
+
},
|
| 1703 |
+
{
|
| 1704 |
+
"epoch": 0.7480190174326465,
|
| 1705 |
+
"grad_norm": 0.2997708530371057,
|
| 1706 |
+
"learning_rate": 3.6203028978741226e-05,
|
| 1707 |
+
"loss": 0.2469,
|
| 1708 |
+
"step": 1180
|
| 1709 |
+
},
|
| 1710 |
+
{
|
| 1711 |
+
"epoch": 0.751188589540412,
|
| 1712 |
+
"grad_norm": 0.32430776300917263,
|
| 1713 |
+
"learning_rate": 3.535451703037626e-05,
|
| 1714 |
+
"loss": 0.2726,
|
| 1715 |
+
"step": 1185
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"epoch": 0.7543581616481775,
|
| 1719 |
+
"grad_norm": 0.2946578935656507,
|
| 1720 |
+
"learning_rate": 3.45139266054715e-05,
|
| 1721 |
+
"loss": 0.2645,
|
| 1722 |
+
"step": 1190
|
| 1723 |
+
},
|
| 1724 |
+
{
|
| 1725 |
+
"epoch": 0.757527733755943,
|
| 1726 |
+
"grad_norm": 0.26638481808591286,
|
| 1727 |
+
"learning_rate": 3.368136070823478e-05,
|
| 1728 |
+
"loss": 0.2465,
|
| 1729 |
+
"step": 1195
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 0.7606973058637084,
|
| 1733 |
+
"grad_norm": 0.3677636374426017,
|
| 1734 |
+
"learning_rate": 3.285692135956515e-05,
|
| 1735 |
+
"loss": 0.3034,
|
| 1736 |
+
"step": 1200
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 0.7606973058637084,
|
| 1740 |
+
"eval_loss": 0.2539891302585602,
|
| 1741 |
+
"eval_runtime": 873.4669,
|
| 1742 |
+
"eval_samples_per_second": 4.579,
|
| 1743 |
+
"eval_steps_per_second": 0.572,
|
| 1744 |
+
"step": 1200
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 0.7638668779714739,
|
| 1748 |
+
"grad_norm": 0.29762017072344943,
|
| 1749 |
+
"learning_rate": 3.2040709584551095e-05,
|
| 1750 |
+
"loss": 0.2547,
|
| 1751 |
+
"step": 1205
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 0.7670364500792393,
|
| 1755 |
+
"grad_norm": 0.35066724794986226,
|
| 1756 |
+
"learning_rate": 3.123282540009139e-05,
|
| 1757 |
+
"loss": 0.3043,
|
| 1758 |
+
"step": 1210
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"epoch": 0.7702060221870047,
|
| 1762 |
+
"grad_norm": 0.27108651599825634,
|
| 1763 |
+
"learning_rate": 3.0433367802639112e-05,
|
| 1764 |
+
"loss": 0.2195,
|
| 1765 |
+
"step": 1215
|
| 1766 |
+
},
|
| 1767 |
+
{
|
| 1768 |
+
"epoch": 0.7733755942947702,
|
| 1769 |
+
"grad_norm": 0.24030479810127725,
|
| 1770 |
+
"learning_rate": 2.9642434756070793e-05,
|
| 1771 |
+
"loss": 0.2545,
|
| 1772 |
+
"step": 1220
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 0.7765451664025357,
|
| 1776 |
+
"grad_norm": 0.288327556838552,
|
| 1777 |
+
"learning_rate": 2.8860123179682242e-05,
|
| 1778 |
+
"loss": 0.2942,
|
| 1779 |
+
"step": 1225
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 0.7797147385103012,
|
| 1783 |
+
"grad_norm": 0.29997783643544385,
|
| 1784 |
+
"learning_rate": 2.8086528936312073e-05,
|
| 1785 |
+
"loss": 0.2407,
|
| 1786 |
+
"step": 1230
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 0.7828843106180665,
|
| 1790 |
+
"grad_norm": 0.2665313932594352,
|
| 1791 |
+
"learning_rate": 2.7321746820595086e-05,
|
| 1792 |
+
"loss": 0.2863,
|
| 1793 |
+
"step": 1235
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 0.786053882725832,
|
| 1797 |
+
"grad_norm": 0.24138106294481415,
|
| 1798 |
+
"learning_rate": 2.6565870547346196e-05,
|
| 1799 |
+
"loss": 0.2443,
|
| 1800 |
+
"step": 1240
|
| 1801 |
+
},
|
| 1802 |
+
{
|
| 1803 |
+
"epoch": 0.7892234548335975,
|
| 1804 |
+
"grad_norm": 0.27410565336257203,
|
| 1805 |
+
"learning_rate": 2.5818992740076873e-05,
|
| 1806 |
+
"loss": 0.2714,
|
| 1807 |
+
"step": 1245
|
| 1808 |
+
},
|
| 1809 |
+
{
|
| 1810 |
+
"epoch": 0.7923930269413629,
|
| 1811 |
+
"grad_norm": 0.3607807135248553,
|
| 1812 |
+
"learning_rate": 2.508120491964512e-05,
|
| 1813 |
+
"loss": 0.3131,
|
| 1814 |
+
"step": 1250
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 0.7955625990491284,
|
| 1818 |
+
"grad_norm": 0.2752324746545014,
|
| 1819 |
+
"learning_rate": 2.435259749304096e-05,
|
| 1820 |
+
"loss": 0.2352,
|
| 1821 |
+
"step": 1255
|
| 1822 |
+
},
|
| 1823 |
+
{
|
| 1824 |
+
"epoch": 0.7987321711568938,
|
| 1825 |
+
"grad_norm": 0.33701412326580854,
|
| 1826 |
+
"learning_rate": 2.3633259742307844e-05,
|
| 1827 |
+
"loss": 0.3121,
|
| 1828 |
+
"step": 1260
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 0.8019017432646592,
|
| 1832 |
+
"grad_norm": 0.2719696587030905,
|
| 1833 |
+
"learning_rate": 2.292327981360245e-05,
|
| 1834 |
+
"loss": 0.2569,
|
| 1835 |
+
"step": 1265
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 0.8050713153724247,
|
| 1839 |
+
"grad_norm": 0.321470064394813,
|
| 1840 |
+
"learning_rate": 2.222274470639324e-05,
|
| 1841 |
+
"loss": 0.2903,
|
| 1842 |
+
"step": 1270
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 0.8082408874801902,
|
| 1846 |
+
"grad_norm": 0.33376441935823614,
|
| 1847 |
+
"learning_rate": 2.1531740262800004e-05,
|
| 1848 |
+
"loss": 0.2712,
|
| 1849 |
+
"step": 1275
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 0.8114104595879557,
|
| 1853 |
+
"grad_norm": 0.3559808478292093,
|
| 1854 |
+
"learning_rate": 2.0850351157074598e-05,
|
| 1855 |
+
"loss": 0.2485,
|
| 1856 |
+
"step": 1280
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 0.8145800316957211,
|
| 1860 |
+
"grad_norm": 0.3006799560470683,
|
| 1861 |
+
"learning_rate": 2.017866088522541e-05,
|
| 1862 |
+
"loss": 0.2735,
|
| 1863 |
+
"step": 1285
|
| 1864 |
+
},
|
| 1865 |
+
{
|
| 1866 |
+
"epoch": 0.8177496038034865,
|
| 1867 |
+
"grad_norm": 0.27868991819615774,
|
| 1868 |
+
"learning_rate": 1.951675175478579e-05,
|
| 1869 |
+
"loss": 0.2479,
|
| 1870 |
+
"step": 1290
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 0.820919175911252,
|
| 1874 |
+
"grad_norm": 0.30796745550467525,
|
| 1875 |
+
"learning_rate": 1.8864704874728346e-05,
|
| 1876 |
+
"loss": 0.2693,
|
| 1877 |
+
"step": 1295
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 0.8240887480190174,
|
| 1881 |
+
"grad_norm": 0.327384705590186,
|
| 1882 |
+
"learning_rate": 1.822260014552587e-05,
|
| 1883 |
+
"loss": 0.2787,
|
| 1884 |
+
"step": 1300
|
| 1885 |
+
},
|
| 1886 |
+
{
|
| 1887 |
+
"epoch": 0.8272583201267829,
|
| 1888 |
+
"grad_norm": 0.2993843751525639,
|
| 1889 |
+
"learning_rate": 1.7590516249360754e-05,
|
| 1890 |
+
"loss": 0.2455,
|
| 1891 |
+
"step": 1305
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 0.8304278922345484,
|
| 1895 |
+
"grad_norm": 0.2979918507317238,
|
| 1896 |
+
"learning_rate": 1.6968530640483127e-05,
|
| 1897 |
+
"loss": 0.2889,
|
| 1898 |
+
"step": 1310
|
| 1899 |
+
},
|
| 1900 |
+
{
|
| 1901 |
+
"epoch": 0.8335974643423137,
|
| 1902 |
+
"grad_norm": 0.2942240760065363,
|
| 1903 |
+
"learning_rate": 1.6356719535720056e-05,
|
| 1904 |
+
"loss": 0.2557,
|
| 1905 |
+
"step": 1315
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 0.8367670364500792,
|
| 1909 |
+
"grad_norm": 0.31698805935759067,
|
| 1910 |
+
"learning_rate": 1.5755157905135843e-05,
|
| 1911 |
+
"loss": 0.2842,
|
| 1912 |
+
"step": 1320
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 0.8399366085578447,
|
| 1916 |
+
"grad_norm": 0.3795639487558114,
|
| 1917 |
+
"learning_rate": 1.5163919462845622e-05,
|
| 1918 |
+
"loss": 0.2979,
|
| 1919 |
+
"step": 1325
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 0.8431061806656102,
|
| 1923 |
+
"grad_norm": 0.2933950396246441,
|
| 1924 |
+
"learning_rate": 1.4583076657982297e-05,
|
| 1925 |
+
"loss": 0.2291,
|
| 1926 |
+
"step": 1330
|
| 1927 |
+
},
|
| 1928 |
+
{
|
| 1929 |
+
"epoch": 0.8462757527733756,
|
| 1930 |
+
"grad_norm": 0.25934135222761445,
|
| 1931 |
+
"learning_rate": 1.401270066581899e-05,
|
| 1932 |
+
"loss": 0.2981,
|
| 1933 |
+
"step": 1335
|
| 1934 |
+
},
|
| 1935 |
+
{
|
| 1936 |
+
"epoch": 0.849445324881141,
|
| 1937 |
+
"grad_norm": 0.2512793866151091,
|
| 1938 |
+
"learning_rate": 1.3452861379047287e-05,
|
| 1939 |
+
"loss": 0.2299,
|
| 1940 |
+
"step": 1340
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 0.8526148969889065,
|
| 1944 |
+
"grad_norm": 0.27890392188122143,
|
| 1945 |
+
"learning_rate": 1.2903627399212747e-05,
|
| 1946 |
+
"loss": 0.2714,
|
| 1947 |
+
"step": 1345
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 0.8557844690966719,
|
| 1951 |
+
"grad_norm": 0.3540435753559853,
|
| 1952 |
+
"learning_rate": 1.2365066028308547e-05,
|
| 1953 |
+
"loss": 0.3208,
|
| 1954 |
+
"step": 1350
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 0.8589540412044374,
|
| 1958 |
+
"grad_norm": 0.3170188652169802,
|
| 1959 |
+
"learning_rate": 1.183724326052854e-05,
|
| 1960 |
+
"loss": 0.261,
|
| 1961 |
+
"step": 1355
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 0.8621236133122029,
|
| 1965 |
+
"grad_norm": 0.287259110452561,
|
| 1966 |
+
"learning_rate": 1.1320223774180428e-05,
|
| 1967 |
+
"loss": 0.2918,
|
| 1968 |
+
"step": 1360
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 0.8652931854199684,
|
| 1972 |
+
"grad_norm": 0.3145063929825825,
|
| 1973 |
+
"learning_rate": 1.0814070923760178e-05,
|
| 1974 |
+
"loss": 0.2562,
|
| 1975 |
+
"step": 1365
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 0.8684627575277337,
|
| 1979 |
+
"grad_norm": 0.29883537499670176,
|
| 1980 |
+
"learning_rate": 1.0318846732188737e-05,
|
| 1981 |
+
"loss": 0.2585,
|
| 1982 |
+
"step": 1370
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 0.8716323296354992,
|
| 1986 |
+
"grad_norm": 0.33602754178177113,
|
| 1987 |
+
"learning_rate": 9.834611883211797e-06,
|
| 1988 |
+
"loss": 0.303,
|
| 1989 |
+
"step": 1375
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 0.8748019017432647,
|
| 1993 |
+
"grad_norm": 0.27917699955310804,
|
| 1994 |
+
"learning_rate": 9.361425713963878e-06,
|
| 1995 |
+
"loss": 0.2399,
|
| 1996 |
+
"step": 1380
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 0.8779714738510301,
|
| 2000 |
+
"grad_norm": 0.29322424380757633,
|
| 2001 |
+
"learning_rate": 8.899346207697134e-06,
|
| 2002 |
+
"loss": 0.3192,
|
| 2003 |
+
"step": 1385
|
| 2004 |
+
},
|
| 2005 |
+
{
|
| 2006 |
+
"epoch": 0.8811410459587956,
|
| 2007 |
+
"grad_norm": 0.32716078301472046,
|
| 2008 |
+
"learning_rate": 8.448429986676298e-06,
|
| 2009 |
+
"loss": 0.256,
|
| 2010 |
+
"step": 1390
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"epoch": 0.884310618066561,
|
| 2014 |
+
"grad_norm": 0.28468261231564157,
|
| 2015 |
+
"learning_rate": 8.00873230524023e-06,
|
| 2016 |
+
"loss": 0.2864,
|
| 2017 |
+
"step": 1395
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 0.8874801901743264,
|
| 2021 |
+
"grad_norm": 0.3481974787604397,
|
| 2022 |
+
"learning_rate": 7.580307043031232e-06,
|
| 2023 |
+
"loss": 0.265,
|
| 2024 |
+
"step": 1400
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 0.8874801901743264,
|
| 2028 |
+
"eval_loss": 0.25099214911460876,
|
| 2029 |
+
"eval_runtime": 873.7854,
|
| 2030 |
+
"eval_samples_per_second": 4.578,
|
| 2031 |
+
"eval_steps_per_second": 0.572,
|
| 2032 |
+
"step": 1400
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.8906497622820919,
|
| 2036 |
+
"grad_norm": 0.2756744352775957,
|
| 2037 |
+
"learning_rate": 7.163206698392744e-06,
|
| 2038 |
+
"loss": 0.2392,
|
| 2039 |
+
"step": 1405
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.8938193343898574,
|
| 2043 |
+
"grad_norm": 0.3070714015760399,
|
| 2044 |
+
"learning_rate": 6.757482381936264e-06,
|
| 2045 |
+
"loss": 0.2722,
|
| 2046 |
+
"step": 1410
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.8969889064976229,
|
| 2050 |
+
"grad_norm": 0.2719682030351016,
|
| 2051 |
+
"learning_rate": 6.36318381027835e-06,
|
| 2052 |
+
"loss": 0.2553,
|
| 2053 |
+
"step": 1415
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.9001584786053882,
|
| 2057 |
+
"grad_norm": 0.30754515515844727,
|
| 2058 |
+
"learning_rate": 5.980359299948568e-06,
|
| 2059 |
+
"loss": 0.2763,
|
| 2060 |
+
"step": 1420
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.9033280507131537,
|
| 2064 |
+
"grad_norm": 0.3599613866897873,
|
| 2065 |
+
"learning_rate": 5.609055761468707e-06,
|
| 2066 |
+
"loss": 0.2987,
|
| 2067 |
+
"step": 1425
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.9064976228209192,
|
| 2071 |
+
"grad_norm": 0.26662442413818216,
|
| 2072 |
+
"learning_rate": 5.249318693604577e-06,
|
| 2073 |
+
"loss": 0.2632,
|
| 2074 |
+
"step": 1430
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.9096671949286846,
|
| 2078 |
+
"grad_norm": 0.2965993748242227,
|
| 2079 |
+
"learning_rate": 4.901192177790692e-06,
|
| 2080 |
+
"loss": 0.2799,
|
| 2081 |
+
"step": 1435
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.9128367670364501,
|
| 2085 |
+
"grad_norm": 0.2923839300339188,
|
| 2086 |
+
"learning_rate": 4.564718872728568e-06,
|
| 2087 |
+
"loss": 0.2464,
|
| 2088 |
+
"step": 1440
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.9160063391442155,
|
| 2092 |
+
"grad_norm": 0.3004256474409844,
|
| 2093 |
+
"learning_rate": 4.2399400091594154e-06,
|
| 2094 |
+
"loss": 0.2775,
|
| 2095 |
+
"step": 1445
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.919175911251981,
|
| 2099 |
+
"grad_norm": 0.30636844288189197,
|
| 2100 |
+
"learning_rate": 3.926895384811835e-06,
|
| 2101 |
+
"loss": 0.2917,
|
| 2102 |
+
"step": 1450
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.9223454833597464,
|
| 2106 |
+
"grad_norm": 0.27018058178290905,
|
| 2107 |
+
"learning_rate": 3.625623359525099e-06,
|
| 2108 |
+
"loss": 0.2522,
|
| 2109 |
+
"step": 1455
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.9255150554675119,
|
| 2113 |
+
"grad_norm": 0.3069766309513976,
|
| 2114 |
+
"learning_rate": 3.33616085054862e-06,
|
| 2115 |
+
"loss": 0.2722,
|
| 2116 |
+
"step": 1460
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.9286846275752774,
|
| 2120 |
+
"grad_norm": 0.2673579253849767,
|
| 2121 |
+
"learning_rate": 3.0585433280180707e-06,
|
| 2122 |
+
"loss": 0.2561,
|
| 2123 |
+
"step": 1465
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.9318541996830428,
|
| 2127 |
+
"grad_norm": 0.2688001276727079,
|
| 2128 |
+
"learning_rate": 2.792804810609173e-06,
|
| 2129 |
+
"loss": 0.2718,
|
| 2130 |
+
"step": 1470
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.9350237717908082,
|
| 2134 |
+
"grad_norm": 0.3331860222359942,
|
| 2135 |
+
"learning_rate": 2.538977861368874e-06,
|
| 2136 |
+
"loss": 0.3163,
|
| 2137 |
+
"step": 1475
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.9381933438985737,
|
| 2141 |
+
"grad_norm": 0.2668325932813764,
|
| 2142 |
+
"learning_rate": 2.2970935837253182e-06,
|
| 2143 |
+
"loss": 0.2393,
|
| 2144 |
+
"step": 1480
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.9413629160063391,
|
| 2148 |
+
"grad_norm": 0.3285498156618503,
|
| 2149 |
+
"learning_rate": 2.0671816176764058e-06,
|
| 2150 |
+
"loss": 0.2862,
|
| 2151 |
+
"step": 1485
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.9445324881141046,
|
| 2155 |
+
"grad_norm": 0.36573862269188245,
|
| 2156 |
+
"learning_rate": 1.8492701361578324e-06,
|
| 2157 |
+
"loss": 0.2447,
|
| 2158 |
+
"step": 1490
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.9477020602218701,
|
| 2162 |
+
"grad_norm": 0.2864139944423568,
|
| 2163 |
+
"learning_rate": 1.6433858415907278e-06,
|
| 2164 |
+
"loss": 0.2777,
|
| 2165 |
+
"step": 1495
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.9508716323296355,
|
| 2169 |
+
"grad_norm": 0.323741034773291,
|
| 2170 |
+
"learning_rate": 1.4495539626097288e-06,
|
| 2171 |
+
"loss": 0.3086,
|
| 2172 |
+
"step": 1500
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.9540412044374009,
|
| 2176 |
+
"grad_norm": 0.2857388007026186,
|
| 2177 |
+
"learning_rate": 1.2677982509714415e-06,
|
| 2178 |
+
"loss": 0.2175,
|
| 2179 |
+
"step": 1505
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.9572107765451664,
|
| 2183 |
+
"grad_norm": 0.2813011213045847,
|
| 2184 |
+
"learning_rate": 1.0981409786439355e-06,
|
| 2185 |
+
"loss": 0.2882,
|
| 2186 |
+
"step": 1510
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.9603803486529319,
|
| 2190 |
+
"grad_norm": 0.27685594779071976,
|
| 2191 |
+
"learning_rate": 9.40602935077639e-07,
|
| 2192 |
+
"loss": 0.23,
|
| 2193 |
+
"step": 1515
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.9635499207606973,
|
| 2197 |
+
"grad_norm": 0.278082958417837,
|
| 2198 |
+
"learning_rate": 7.952034246577977e-07,
|
| 2199 |
+
"loss": 0.2814,
|
| 2200 |
+
"step": 1520
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.9667194928684627,
|
| 2204 |
+
"grad_norm": 0.332411253150925,
|
| 2205 |
+
"learning_rate": 6.619602643389899e-07,
|
| 2206 |
+
"loss": 0.2772,
|
| 2207 |
+
"step": 1525
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.9698890649762282,
|
| 2211 |
+
"grad_norm": 0.28541188324654354,
|
| 2212 |
+
"learning_rate": 5.408897814618175e-07,
|
| 2213 |
+
"loss": 0.2456,
|
| 2214 |
+
"step": 1530
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.9730586370839936,
|
| 2218 |
+
"grad_norm": 0.289051402982161,
|
| 2219 |
+
"learning_rate": 4.320068117522835e-07,
|
| 2220 |
+
"loss": 0.2659,
|
| 2221 |
+
"step": 1535
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.9762282091917591,
|
| 2225 |
+
"grad_norm": 0.2896831822321737,
|
| 2226 |
+
"learning_rate": 3.35324697503725e-07,
|
| 2227 |
+
"loss": 0.2721,
|
| 2228 |
+
"step": 1540
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.9793977812995246,
|
| 2232 |
+
"grad_norm": 0.31950029347694936,
|
| 2233 |
+
"learning_rate": 2.508552859419977e-07,
|
| 2234 |
+
"loss": 0.2622,
|
| 2235 |
+
"step": 1545
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.9825673534072901,
|
| 2239 |
+
"grad_norm": 0.33661523682392047,
|
| 2240 |
+
"learning_rate": 1.7860892777367133e-07,
|
| 2241 |
+
"loss": 0.2731,
|
| 2242 |
+
"step": 1550
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.9857369255150554,
|
| 2246 |
+
"grad_norm": 0.2522879758084615,
|
| 2247 |
+
"learning_rate": 1.1859447591769934e-07,
|
| 2248 |
+
"loss": 0.2291,
|
| 2249 |
+
"step": 1555
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.9889064976228209,
|
| 2253 |
+
"grad_norm": 0.2923729662272973,
|
| 2254 |
+
"learning_rate": 7.081928442057573e-08,
|
| 2255 |
+
"loss": 0.2972,
|
| 2256 |
+
"step": 1560
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.9920760697305864,
|
| 2260 |
+
"grad_norm": 0.24814229174923821,
|
| 2261 |
+
"learning_rate": 3.5289207555233573e-08,
|
| 2262 |
+
"loss": 0.2586,
|
| 2263 |
+
"step": 1565
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.9952456418383518,
|
| 2267 |
+
"grad_norm": 0.24322900794711846,
|
| 2268 |
+
"learning_rate": 1.2008599103618956e-08,
|
| 2269 |
+
"loss": 0.2751,
|
| 2270 |
+
"step": 1570
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.9984152139461173,
|
| 2274 |
+
"grad_norm": 0.4374899080765362,
|
| 2275 |
+
"learning_rate": 9.803118232398768e-10,
|
| 2276 |
+
"loss": 0.2981,
|
| 2277 |
+
"step": 1575
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.9996830427892235,
|
| 2281 |
+
"step": 1577,
|
| 2282 |
+
"total_flos": 8013042675351552.0,
|
| 2283 |
+
"train_loss": 0.3010184336819827,
|
| 2284 |
+
"train_runtime": 18281.3669,
|
| 2285 |
+
"train_samples_per_second": 1.381,
|
| 2286 |
+
"train_steps_per_second": 0.086
|
| 2287 |
+
}
|
| 2288 |
+
],
|
| 2289 |
+
"logging_steps": 5,
|
| 2290 |
+
"max_steps": 1577,
|
| 2291 |
+
"num_input_tokens_seen": 0,
|
| 2292 |
+
"num_train_epochs": 1,
|
| 2293 |
+
"save_steps": 200,
|
| 2294 |
+
"stateful_callbacks": {
|
| 2295 |
+
"TrainerControl": {
|
| 2296 |
+
"args": {
|
| 2297 |
+
"should_epoch_stop": false,
|
| 2298 |
+
"should_evaluate": false,
|
| 2299 |
+
"should_log": false,
|
| 2300 |
+
"should_save": true,
|
| 2301 |
+
"should_training_stop": true
|
| 2302 |
+
},
|
| 2303 |
+
"attributes": {}
|
| 2304 |
+
}
|
| 2305 |
+
},
|
| 2306 |
+
"total_flos": 8013042675351552.0,
|
| 2307 |
+
"train_batch_size": 2,
|
| 2308 |
+
"trial_name": null,
|
| 2309 |
+
"trial_params": null
|
| 2310 |
+
}
|