Model save
Browse files- README.md +67 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- train_results.json +9 -0
- trainer_state.json +1912 -0
README.md
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- generator
|
| 10 |
+
model-index:
|
| 11 |
+
- name: prometheus-7b-preference-0.1p-rare-seed42
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# prometheus-7b-preference-0.1p-rare-seed42
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the generator dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.4708
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 1e-05
|
| 42 |
+
- train_batch_size: 2
|
| 43 |
+
- eval_batch_size: 1
|
| 44 |
+
- seed: 42
|
| 45 |
+
- distributed_type: multi-GPU
|
| 46 |
+
- num_devices: 4
|
| 47 |
+
- gradient_accumulation_steps: 4
|
| 48 |
+
- total_train_batch_size: 32
|
| 49 |
+
- total_eval_batch_size: 4
|
| 50 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 51 |
+
- lr_scheduler_type: cosine
|
| 52 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 53 |
+
- num_epochs: 1
|
| 54 |
+
|
| 55 |
+
### Training results
|
| 56 |
+
|
| 57 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 58 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 59 |
+
| 0.4741 | 0.9994 | 1325 | 0.4708 |
|
| 60 |
+
|
| 61 |
+
|
| 62 |
+
### Framework versions
|
| 63 |
+
|
| 64 |
+
- Transformers 4.43.4
|
| 65 |
+
- Pytorch 2.3.1+cu121
|
| 66 |
+
- Datasets 2.19.1
|
| 67 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9994342824816141,
|
| 3 |
+
"total_flos": 277375698862080.0,
|
| 4 |
+
"train_loss": 0.6321603435840247,
|
| 5 |
+
"train_runtime": 30135.1647,
|
| 6 |
+
"train_samples": 108769,
|
| 7 |
+
"train_samples_per_second": 1.408,
|
| 8 |
+
"train_steps_per_second": 0.044
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.43.4"
|
| 6 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9994342824816141,
|
| 3 |
+
"total_flos": 277375698862080.0,
|
| 4 |
+
"train_loss": 0.6321603435840247,
|
| 5 |
+
"train_runtime": 30135.1647,
|
| 6 |
+
"train_samples": 108769,
|
| 7 |
+
"train_samples_per_second": 1.408,
|
| 8 |
+
"train_steps_per_second": 0.044
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1912 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9994342824816141,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1325,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0007542900245144258,
|
| 13 |
+
"grad_norm": 24.0743322267596,
|
| 14 |
+
"learning_rate": 7.518796992481203e-08,
|
| 15 |
+
"loss": 1.4329,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.003771450122572129,
|
| 20 |
+
"grad_norm": 24.762983495374982,
|
| 21 |
+
"learning_rate": 3.7593984962406015e-07,
|
| 22 |
+
"loss": 1.417,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.007542900245144258,
|
| 27 |
+
"grad_norm": 8.686384748409985,
|
| 28 |
+
"learning_rate": 7.518796992481203e-07,
|
| 29 |
+
"loss": 1.3206,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.011314350367716388,
|
| 34 |
+
"grad_norm": 10.36105058434612,
|
| 35 |
+
"learning_rate": 1.1278195488721805e-06,
|
| 36 |
+
"loss": 1.1564,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.015085800490288516,
|
| 41 |
+
"grad_norm": 3.2096721215381483,
|
| 42 |
+
"learning_rate": 1.5037593984962406e-06,
|
| 43 |
+
"loss": 1.0254,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.018857250612860643,
|
| 48 |
+
"grad_norm": 2.8041993817362436,
|
| 49 |
+
"learning_rate": 1.8796992481203007e-06,
|
| 50 |
+
"loss": 0.9632,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.022628700735432775,
|
| 55 |
+
"grad_norm": 2.4316167661473758,
|
| 56 |
+
"learning_rate": 2.255639097744361e-06,
|
| 57 |
+
"loss": 0.9285,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.026400150858004904,
|
| 62 |
+
"grad_norm": 2.347169321112029,
|
| 63 |
+
"learning_rate": 2.631578947368421e-06,
|
| 64 |
+
"loss": 0.9228,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.030171600980577033,
|
| 69 |
+
"grad_norm": 2.23082557930139,
|
| 70 |
+
"learning_rate": 3.007518796992481e-06,
|
| 71 |
+
"loss": 0.8955,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03394305110314916,
|
| 76 |
+
"grad_norm": 2.3187330629715484,
|
| 77 |
+
"learning_rate": 3.3834586466165413e-06,
|
| 78 |
+
"loss": 0.9049,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.037714501225721286,
|
| 83 |
+
"grad_norm": 2.239439563631581,
|
| 84 |
+
"learning_rate": 3.7593984962406014e-06,
|
| 85 |
+
"loss": 0.894,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04148595134829342,
|
| 90 |
+
"grad_norm": 2.3025921470010267,
|
| 91 |
+
"learning_rate": 4.135338345864662e-06,
|
| 92 |
+
"loss": 0.8807,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04525740147086555,
|
| 97 |
+
"grad_norm": 2.387099716939349,
|
| 98 |
+
"learning_rate": 4.511278195488722e-06,
|
| 99 |
+
"loss": 0.875,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.04902885159343768,
|
| 104 |
+
"grad_norm": 2.3813541654085046,
|
| 105 |
+
"learning_rate": 4.887218045112782e-06,
|
| 106 |
+
"loss": 0.8674,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.05280030171600981,
|
| 111 |
+
"grad_norm": 2.609644080569622,
|
| 112 |
+
"learning_rate": 5.263157894736842e-06,
|
| 113 |
+
"loss": 0.8655,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.056571751838581937,
|
| 118 |
+
"grad_norm": 2.571607734979684,
|
| 119 |
+
"learning_rate": 5.639097744360903e-06,
|
| 120 |
+
"loss": 0.862,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.060343201961154065,
|
| 125 |
+
"grad_norm": 2.2582812119753903,
|
| 126 |
+
"learning_rate": 6.015037593984962e-06,
|
| 127 |
+
"loss": 0.8518,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06411465208372619,
|
| 132 |
+
"grad_norm": 2.450433458259798,
|
| 133 |
+
"learning_rate": 6.390977443609023e-06,
|
| 134 |
+
"loss": 0.856,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.06788610220629832,
|
| 139 |
+
"grad_norm": 2.58014673334683,
|
| 140 |
+
"learning_rate": 6.766917293233083e-06,
|
| 141 |
+
"loss": 0.8528,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07165755232887044,
|
| 146 |
+
"grad_norm": 2.3740208001929743,
|
| 147 |
+
"learning_rate": 7.1428571428571436e-06,
|
| 148 |
+
"loss": 0.8331,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.07542900245144257,
|
| 153 |
+
"grad_norm": 2.413650287157212,
|
| 154 |
+
"learning_rate": 7.518796992481203e-06,
|
| 155 |
+
"loss": 0.8605,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.07920045257401472,
|
| 160 |
+
"grad_norm": 2.355976041456668,
|
| 161 |
+
"learning_rate": 7.894736842105265e-06,
|
| 162 |
+
"loss": 0.8389,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08297190269658684,
|
| 167 |
+
"grad_norm": 2.3146510995724334,
|
| 168 |
+
"learning_rate": 8.270676691729324e-06,
|
| 169 |
+
"loss": 0.8398,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.08674335281915897,
|
| 174 |
+
"grad_norm": 2.539423496991191,
|
| 175 |
+
"learning_rate": 8.646616541353385e-06,
|
| 176 |
+
"loss": 0.8281,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.0905148029417311,
|
| 181 |
+
"grad_norm": 2.3110234365436115,
|
| 182 |
+
"learning_rate": 9.022556390977444e-06,
|
| 183 |
+
"loss": 0.8355,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.09428625306430323,
|
| 188 |
+
"grad_norm": 2.3074854008707644,
|
| 189 |
+
"learning_rate": 9.398496240601505e-06,
|
| 190 |
+
"loss": 0.8378,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.09805770318687536,
|
| 195 |
+
"grad_norm": 2.198478602649139,
|
| 196 |
+
"learning_rate": 9.774436090225564e-06,
|
| 197 |
+
"loss": 0.8255,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.10182915330944749,
|
| 202 |
+
"grad_norm": 2.372634478935117,
|
| 203 |
+
"learning_rate": 9.999930538168684e-06,
|
| 204 |
+
"loss": 0.8152,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.10560060343201962,
|
| 209 |
+
"grad_norm": 2.3429810617684406,
|
| 210 |
+
"learning_rate": 9.999149114730812e-06,
|
| 211 |
+
"loss": 0.829,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.10937205355459174,
|
| 216 |
+
"grad_norm": 2.2344447897248783,
|
| 217 |
+
"learning_rate": 9.997499576714369e-06,
|
| 218 |
+
"loss": 0.8299,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.11314350367716387,
|
| 223 |
+
"grad_norm": 2.241744206103708,
|
| 224 |
+
"learning_rate": 9.9949822105657e-06,
|
| 225 |
+
"loss": 0.8253,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.116914953799736,
|
| 230 |
+
"grad_norm": 2.2980391560118862,
|
| 231 |
+
"learning_rate": 9.991597453431652e-06,
|
| 232 |
+
"loss": 0.8258,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.12068640392230813,
|
| 237 |
+
"grad_norm": 2.501411375935894,
|
| 238 |
+
"learning_rate": 9.987345893083652e-06,
|
| 239 |
+
"loss": 0.8165,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.12445785404488026,
|
| 244 |
+
"grad_norm": 2.286238996613632,
|
| 245 |
+
"learning_rate": 9.982228267815644e-06,
|
| 246 |
+
"loss": 0.8128,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.12822930416745237,
|
| 251 |
+
"grad_norm": 2.179417789665571,
|
| 252 |
+
"learning_rate": 9.97624546631588e-06,
|
| 253 |
+
"loss": 0.8193,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.1320007542900245,
|
| 258 |
+
"grad_norm": 2.186299499644603,
|
| 259 |
+
"learning_rate": 9.969398527512607e-06,
|
| 260 |
+
"loss": 0.8153,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.13577220441259663,
|
| 265 |
+
"grad_norm": 2.177003107850399,
|
| 266 |
+
"learning_rate": 9.96168864039364e-06,
|
| 267 |
+
"loss": 0.8004,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.13954365453516876,
|
| 272 |
+
"grad_norm": 2.5007296087774575,
|
| 273 |
+
"learning_rate": 9.9531171437999e-06,
|
| 274 |
+
"loss": 0.8057,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.1433151046577409,
|
| 279 |
+
"grad_norm": 2.2265359668215696,
|
| 280 |
+
"learning_rate": 9.943685526192921e-06,
|
| 281 |
+
"loss": 0.8045,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.14708655478031302,
|
| 286 |
+
"grad_norm": 2.1009649317490116,
|
| 287 |
+
"learning_rate": 9.93339542539637e-06,
|
| 288 |
+
"loss": 0.7964,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.15085800490288515,
|
| 293 |
+
"grad_norm": 2.0704593465422008,
|
| 294 |
+
"learning_rate": 9.922248628311641e-06,
|
| 295 |
+
"loss": 0.8007,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.1546294550254573,
|
| 300 |
+
"grad_norm": 2.2278475317912694,
|
| 301 |
+
"learning_rate": 9.91024707060755e-06,
|
| 302 |
+
"loss": 0.8074,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.15840090514802943,
|
| 307 |
+
"grad_norm": 2.3927523335337293,
|
| 308 |
+
"learning_rate": 9.89739283638421e-06,
|
| 309 |
+
"loss": 0.7964,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.16217235527060156,
|
| 314 |
+
"grad_norm": 2.2597054171285826,
|
| 315 |
+
"learning_rate": 9.883688157811102e-06,
|
| 316 |
+
"loss": 0.7925,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.1659438053931737,
|
| 321 |
+
"grad_norm": 2.2971840831123806,
|
| 322 |
+
"learning_rate": 9.86913541473948e-06,
|
| 323 |
+
"loss": 0.7928,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.16971525551574582,
|
| 328 |
+
"grad_norm": 2.161562265562174,
|
| 329 |
+
"learning_rate": 9.853737134289086e-06,
|
| 330 |
+
"loss": 0.78,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.17348670563831795,
|
| 335 |
+
"grad_norm": 2.087919478225615,
|
| 336 |
+
"learning_rate": 9.83749599040931e-06,
|
| 337 |
+
"loss": 0.7852,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.17725815576089007,
|
| 342 |
+
"grad_norm": 2.1464178337018422,
|
| 343 |
+
"learning_rate": 9.820414803414862e-06,
|
| 344 |
+
"loss": 0.7864,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.1810296058834622,
|
| 349 |
+
"grad_norm": 2.1956780101934115,
|
| 350 |
+
"learning_rate": 9.802496539496008e-06,
|
| 351 |
+
"loss": 0.777,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.18480105600603433,
|
| 356 |
+
"grad_norm": 2.2171827211629456,
|
| 357 |
+
"learning_rate": 9.783744310203492e-06,
|
| 358 |
+
"loss": 0.7738,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.18857250612860646,
|
| 363 |
+
"grad_norm": 2.1354704419052,
|
| 364 |
+
"learning_rate": 9.764161371908195e-06,
|
| 365 |
+
"loss": 0.7833,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1923439562511786,
|
| 370 |
+
"grad_norm": 2.244895598639118,
|
| 371 |
+
"learning_rate": 9.743751125235674e-06,
|
| 372 |
+
"loss": 0.7832,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.19611540637375072,
|
| 377 |
+
"grad_norm": 2.1690612043700983,
|
| 378 |
+
"learning_rate": 9.722517114475622e-06,
|
| 379 |
+
"loss": 0.7759,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.19988685649632285,
|
| 384 |
+
"grad_norm": 2.465068879237022,
|
| 385 |
+
"learning_rate": 9.7004630269664e-06,
|
| 386 |
+
"loss": 0.7736,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.20365830661889497,
|
| 391 |
+
"grad_norm": 2.3445933624332973,
|
| 392 |
+
"learning_rate": 9.67759269245472e-06,
|
| 393 |
+
"loss": 0.7617,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2074297567414671,
|
| 398 |
+
"grad_norm": 2.131498314204489,
|
| 399 |
+
"learning_rate": 9.6539100824306e-06,
|
| 400 |
+
"loss": 0.7616,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.21120120686403923,
|
| 405 |
+
"grad_norm": 2.365920326550214,
|
| 406 |
+
"learning_rate": 9.629419309437704e-06,
|
| 407 |
+
"loss": 0.7555,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.21497265698661136,
|
| 412 |
+
"grad_norm": 2.330336022690634,
|
| 413 |
+
"learning_rate": 9.60412462635919e-06,
|
| 414 |
+
"loss": 0.7541,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.2187441071091835,
|
| 419 |
+
"grad_norm": 2.2560710591306634,
|
| 420 |
+
"learning_rate": 9.578030425679188e-06,
|
| 421 |
+
"loss": 0.751,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.22251555723175562,
|
| 426 |
+
"grad_norm": 2.054455615320725,
|
| 427 |
+
"learning_rate": 9.551141238720027e-06,
|
| 428 |
+
"loss": 0.7491,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.22628700735432775,
|
| 433 |
+
"grad_norm": 2.3326134119870465,
|
| 434 |
+
"learning_rate": 9.523461734855373e-06,
|
| 435 |
+
"loss": 0.7494,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.23005845747689987,
|
| 440 |
+
"grad_norm": 2.0451775689136995,
|
| 441 |
+
"learning_rate": 9.494996720699363e-06,
|
| 442 |
+
"loss": 0.7527,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.233829907599472,
|
| 447 |
+
"grad_norm": 2.2108967714871275,
|
| 448 |
+
"learning_rate": 9.465751139271943e-06,
|
| 449 |
+
"loss": 0.751,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.23760135772204413,
|
| 454 |
+
"grad_norm": 2.09565099808324,
|
| 455 |
+
"learning_rate": 9.435730069140486e-06,
|
| 456 |
+
"loss": 0.741,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.24137280784461626,
|
| 461 |
+
"grad_norm": 2.1588491738436937,
|
| 462 |
+
"learning_rate": 9.404938723537894e-06,
|
| 463 |
+
"loss": 0.7429,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.2451442579671884,
|
| 468 |
+
"grad_norm": 2.056502617886576,
|
| 469 |
+
"learning_rate": 9.373382449457305e-06,
|
| 470 |
+
"loss": 0.7341,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.24891570808976052,
|
| 475 |
+
"grad_norm": 2.067525516090952,
|
| 476 |
+
"learning_rate": 9.341066726723574e-06,
|
| 477 |
+
"loss": 0.7522,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.2526871582123326,
|
| 482 |
+
"grad_norm": 2.141985494588096,
|
| 483 |
+
"learning_rate": 9.307997167041688e-06,
|
| 484 |
+
"loss": 0.736,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.25645860833490475,
|
| 489 |
+
"grad_norm": 2.175709280838213,
|
| 490 |
+
"learning_rate": 9.274179513022284e-06,
|
| 491 |
+
"loss": 0.7427,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.2602300584574769,
|
| 496 |
+
"grad_norm": 2.661524805942838,
|
| 497 |
+
"learning_rate": 9.23961963718442e-06,
|
| 498 |
+
"loss": 0.7354,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.264001508580049,
|
| 503 |
+
"grad_norm": 2.3806741154817606,
|
| 504 |
+
"learning_rate": 9.204323540935808e-06,
|
| 505 |
+
"loss": 0.7171,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.26777295870262113,
|
| 510 |
+
"grad_norm": 2.2844034910230384,
|
| 511 |
+
"learning_rate": 9.168297353530654e-06,
|
| 512 |
+
"loss": 0.7246,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.27154440882519326,
|
| 517 |
+
"grad_norm": 2.2290764994416756,
|
| 518 |
+
"learning_rate": 9.131547331005288e-06,
|
| 519 |
+
"loss": 0.7219,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.2753158589477654,
|
| 524 |
+
"grad_norm": 2.1718562436320727,
|
| 525 |
+
"learning_rate": 9.094079855091797e-06,
|
| 526 |
+
"loss": 0.7106,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.2790873090703375,
|
| 531 |
+
"grad_norm": 2.055976595872544,
|
| 532 |
+
"learning_rate": 9.055901432109821e-06,
|
| 533 |
+
"loss": 0.7242,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.28285875919290965,
|
| 538 |
+
"grad_norm": 2.2447216963743575,
|
| 539 |
+
"learning_rate": 9.017018691836714e-06,
|
| 540 |
+
"loss": 0.7269,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.2866302093154818,
|
| 545 |
+
"grad_norm": 2.108026977341414,
|
| 546 |
+
"learning_rate": 8.97743838635626e-06,
|
| 547 |
+
"loss": 0.7253,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.2904016594380539,
|
| 552 |
+
"grad_norm": 2.149969361483023,
|
| 553 |
+
"learning_rate": 8.937167388886163e-06,
|
| 554 |
+
"loss": 0.7154,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.29417310956062603,
|
| 559 |
+
"grad_norm": 2.1649261230436543,
|
| 560 |
+
"learning_rate": 8.896212692584503e-06,
|
| 561 |
+
"loss": 0.7215,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.29794455968319816,
|
| 566 |
+
"grad_norm": 1.9995253010334968,
|
| 567 |
+
"learning_rate": 8.854581409335341e-06,
|
| 568 |
+
"loss": 0.7148,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.3017160098057703,
|
| 573 |
+
"grad_norm": 2.2916154939190783,
|
| 574 |
+
"learning_rate": 8.812280768513732e-06,
|
| 575 |
+
"loss": 0.7089,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.3054874599283425,
|
| 580 |
+
"grad_norm": 2.1141628178573466,
|
| 581 |
+
"learning_rate": 8.76931811573033e-06,
|
| 582 |
+
"loss": 0.708,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3092589100509146,
|
| 587 |
+
"grad_norm": 2.175109844911728,
|
| 588 |
+
"learning_rate": 8.725700911555792e-06,
|
| 589 |
+
"loss": 0.7037,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.31303036017348673,
|
| 594 |
+
"grad_norm": 2.101023552691967,
|
| 595 |
+
"learning_rate": 8.681436730225245e-06,
|
| 596 |
+
"loss": 0.7053,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.31680181029605886,
|
| 601 |
+
"grad_norm": 2.101523445074263,
|
| 602 |
+
"learning_rate": 8.636533258322995e-06,
|
| 603 |
+
"loss": 0.6861,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.320573260418631,
|
| 608 |
+
"grad_norm": 5.819641297709661,
|
| 609 |
+
"learning_rate": 8.590998293447728e-06,
|
| 610 |
+
"loss": 0.6921,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.3243447105412031,
|
| 615 |
+
"grad_norm": 2.3399412061978784,
|
| 616 |
+
"learning_rate": 8.544839742858449e-06,
|
| 617 |
+
"loss": 0.691,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.32811616066377525,
|
| 622 |
+
"grad_norm": 2.2092909626343173,
|
| 623 |
+
"learning_rate": 8.49806562210136e-06,
|
| 624 |
+
"loss": 0.6934,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.3318876107863474,
|
| 629 |
+
"grad_norm": 2.0855142628275374,
|
| 630 |
+
"learning_rate": 8.450684053617936e-06,
|
| 631 |
+
"loss": 0.693,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.3356590609089195,
|
| 636 |
+
"grad_norm": 2.259615514364146,
|
| 637 |
+
"learning_rate": 8.402703265334455e-06,
|
| 638 |
+
"loss": 0.6949,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.33943051103149163,
|
| 643 |
+
"grad_norm": 2.176855477838487,
|
| 644 |
+
"learning_rate": 8.35413158923319e-06,
|
| 645 |
+
"loss": 0.6943,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.34320196115406376,
|
| 650 |
+
"grad_norm": 5.3013314444050295,
|
| 651 |
+
"learning_rate": 8.304977459905539e-06,
|
| 652 |
+
"loss": 0.693,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.3469734112766359,
|
| 657 |
+
"grad_norm": 2.3136109012385853,
|
| 658 |
+
"learning_rate": 8.255249413087342e-06,
|
| 659 |
+
"loss": 0.6839,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.350744861399208,
|
| 664 |
+
"grad_norm": 2.5530506882002775,
|
| 665 |
+
"learning_rate": 8.20495608417663e-06,
|
| 666 |
+
"loss": 0.6837,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.35451631152178015,
|
| 671 |
+
"grad_norm": 2.2039921142230923,
|
| 672 |
+
"learning_rate": 8.154106206734056e-06,
|
| 673 |
+
"loss": 0.6757,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.3582877616443523,
|
| 678 |
+
"grad_norm": 2.0873079113549005,
|
| 679 |
+
"learning_rate": 8.102708610966306e-06,
|
| 680 |
+
"loss": 0.6845,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3620592117669244,
|
| 685 |
+
"grad_norm": 2.2652146917109994,
|
| 686 |
+
"learning_rate": 8.050772222192703e-06,
|
| 687 |
+
"loss": 0.679,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.36583066188949653,
|
| 692 |
+
"grad_norm": 2.342496143877203,
|
| 693 |
+
"learning_rate": 7.998306059295302e-06,
|
| 694 |
+
"loss": 0.6763,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.36960211201206866,
|
| 699 |
+
"grad_norm": 2.1611026997479246,
|
| 700 |
+
"learning_rate": 7.945319233152748e-06,
|
| 701 |
+
"loss": 0.6731,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.3733735621346408,
|
| 706 |
+
"grad_norm": 2.1522125366560556,
|
| 707 |
+
"learning_rate": 7.891820945058137e-06,
|
| 708 |
+
"loss": 0.6704,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3771450122572129,
|
| 713 |
+
"grad_norm": 2.225656595731522,
|
| 714 |
+
"learning_rate": 7.837820485121196e-06,
|
| 715 |
+
"loss": 0.6577,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.38091646237978505,
|
| 720 |
+
"grad_norm": 2.05947893221226,
|
| 721 |
+
"learning_rate": 7.783327230655036e-06,
|
| 722 |
+
"loss": 0.657,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3846879125023572,
|
| 727 |
+
"grad_norm": 2.0489553480122553,
|
| 728 |
+
"learning_rate": 7.728350644547748e-06,
|
| 729 |
+
"loss": 0.6599,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.3884593626249293,
|
| 734 |
+
"grad_norm": 2.3536125168662134,
|
| 735 |
+
"learning_rate": 7.672900273619158e-06,
|
| 736 |
+
"loss": 0.6379,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.39223081274750143,
|
| 741 |
+
"grad_norm": 2.278757928608761,
|
| 742 |
+
"learning_rate": 7.616985746962995e-06,
|
| 743 |
+
"loss": 0.6584,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.39600226287007356,
|
| 748 |
+
"grad_norm": 2.2890835012690385,
|
| 749 |
+
"learning_rate": 7.560616774274775e-06,
|
| 750 |
+
"loss": 0.6675,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.3997737129926457,
|
| 755 |
+
"grad_norm": 2.052679763492931,
|
| 756 |
+
"learning_rate": 7.503803144165691e-06,
|
| 757 |
+
"loss": 0.6532,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.4035451631152178,
|
| 762 |
+
"grad_norm": 2.1962590249997804,
|
| 763 |
+
"learning_rate": 7.446554722462789e-06,
|
| 764 |
+
"loss": 0.6452,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.40731661323778995,
|
| 769 |
+
"grad_norm": 2.1492418404505513,
|
| 770 |
+
"learning_rate": 7.388881450495753e-06,
|
| 771 |
+
"loss": 0.6399,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.4110880633603621,
|
| 776 |
+
"grad_norm": 2.140785983967721,
|
| 777 |
+
"learning_rate": 7.33079334337056e-06,
|
| 778 |
+
"loss": 0.6519,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.4148595134829342,
|
| 783 |
+
"grad_norm": 2.2023514849237857,
|
| 784 |
+
"learning_rate": 7.272300488230342e-06,
|
| 785 |
+
"loss": 0.6451,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.41863096360550633,
|
| 790 |
+
"grad_norm": 2.1139769229137197,
|
| 791 |
+
"learning_rate": 7.213413042503714e-06,
|
| 792 |
+
"loss": 0.6465,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.42240241372807846,
|
| 797 |
+
"grad_norm": 2.1321527320416216,
|
| 798 |
+
"learning_rate": 7.154141232140925e-06,
|
| 799 |
+
"loss": 0.632,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.4261738638506506,
|
| 804 |
+
"grad_norm": 2.1408532489827383,
|
| 805 |
+
"learning_rate": 7.094495349838093e-06,
|
| 806 |
+
"loss": 0.64,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.4299453139732227,
|
| 811 |
+
"grad_norm": 2.1424271774917343,
|
| 812 |
+
"learning_rate": 7.034485753249846e-06,
|
| 813 |
+
"loss": 0.6318,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.43371676409579485,
|
| 818 |
+
"grad_norm": 2.4132135353951223,
|
| 819 |
+
"learning_rate": 6.9741228631907e-06,
|
| 820 |
+
"loss": 0.6388,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.437488214218367,
|
| 825 |
+
"grad_norm": 2.2228430360361373,
|
| 826 |
+
"learning_rate": 6.913417161825449e-06,
|
| 827 |
+
"loss": 0.6232,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.4412596643409391,
|
| 832 |
+
"grad_norm": 2.101068755168653,
|
| 833 |
+
"learning_rate": 6.852379190848923e-06,
|
| 834 |
+
"loss": 0.6274,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.44503111446351123,
|
| 839 |
+
"grad_norm": 2.283059476668117,
|
| 840 |
+
"learning_rate": 6.791019549655384e-06,
|
| 841 |
+
"loss": 0.6178,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.44880256458608336,
|
| 846 |
+
"grad_norm": 2.1112204908741137,
|
| 847 |
+
"learning_rate": 6.72934889349792e-06,
|
| 848 |
+
"loss": 0.631,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.4525740147086555,
|
| 853 |
+
"grad_norm": 2.1049933744197813,
|
| 854 |
+
"learning_rate": 6.667377931638144e-06,
|
| 855 |
+
"loss": 0.6134,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.4563454648312276,
|
| 860 |
+
"grad_norm": 2.235411557732047,
|
| 861 |
+
"learning_rate": 6.605117425486483e-06,
|
| 862 |
+
"loss": 0.6119,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.46011691495379975,
|
| 867 |
+
"grad_norm": 2.0415112264754107,
|
| 868 |
+
"learning_rate": 6.542578186733453e-06,
|
| 869 |
+
"loss": 0.6173,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.4638883650763719,
|
| 874 |
+
"grad_norm": 2.185203312850119,
|
| 875 |
+
"learning_rate": 6.47977107547217e-06,
|
| 876 |
+
"loss": 0.6219,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.467659815198944,
|
| 881 |
+
"grad_norm": 2.21664954602057,
|
| 882 |
+
"learning_rate": 6.4167069983124765e-06,
|
| 883 |
+
"loss": 0.6144,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.47143126532151614,
|
| 888 |
+
"grad_norm": 2.1363794732205807,
|
| 889 |
+
"learning_rate": 6.353396906486971e-06,
|
| 890 |
+
"loss": 0.621,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.47520271544408826,
|
| 895 |
+
"grad_norm": 2.2099254913643502,
|
| 896 |
+
"learning_rate": 6.289851793949311e-06,
|
| 897 |
+
"loss": 0.6118,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.4789741655666604,
|
| 902 |
+
"grad_norm": 2.3004798131436055,
|
| 903 |
+
"learning_rate": 6.2260826954650795e-06,
|
| 904 |
+
"loss": 0.6197,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.4827456156892325,
|
| 909 |
+
"grad_norm": 2.154968570593537,
|
| 910 |
+
"learning_rate": 6.1621006846955676e-06,
|
| 911 |
+
"loss": 0.6012,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.48651706581180465,
|
| 916 |
+
"grad_norm": 2.2605674779702847,
|
| 917 |
+
"learning_rate": 6.097916872274815e-06,
|
| 918 |
+
"loss": 0.601,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.4902885159343768,
|
| 923 |
+
"grad_norm": 2.2052321176133134,
|
| 924 |
+
"learning_rate": 6.033542403880216e-06,
|
| 925 |
+
"loss": 0.6092,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.4940599660569489,
|
| 930 |
+
"grad_norm": 2.1808520045523427,
|
| 931 |
+
"learning_rate": 5.96898845829704e-06,
|
| 932 |
+
"loss": 0.6059,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.49783141617952104,
|
| 937 |
+
"grad_norm": 2.300580579687932,
|
| 938 |
+
"learning_rate": 5.904266245477219e-06,
|
| 939 |
+
"loss": 0.5946,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.5016028663020932,
|
| 944 |
+
"grad_norm": 2.0297693933559398,
|
| 945 |
+
"learning_rate": 5.839387004592705e-06,
|
| 946 |
+
"loss": 0.5859,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.5053743164246652,
|
| 951 |
+
"grad_norm": 2.0914074101770623,
|
| 952 |
+
"learning_rate": 5.7743620020837565e-06,
|
| 953 |
+
"loss": 0.598,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.5091457665472374,
|
| 958 |
+
"grad_norm": 2.0766975702892836,
|
| 959 |
+
"learning_rate": 5.709202529702501e-06,
|
| 960 |
+
"loss": 0.5879,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.5129172166698095,
|
| 965 |
+
"grad_norm": 2.213588628181455,
|
| 966 |
+
"learning_rate": 5.643919902552091e-06,
|
| 967 |
+
"loss": 0.5913,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.5166886667923817,
|
| 972 |
+
"grad_norm": 2.1397173162628156,
|
| 973 |
+
"learning_rate": 5.578525457121807e-06,
|
| 974 |
+
"loss": 0.583,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.5204601169149538,
|
| 979 |
+
"grad_norm": 2.2408994529326103,
|
| 980 |
+
"learning_rate": 5.513030549318453e-06,
|
| 981 |
+
"loss": 0.5783,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.5242315670375259,
|
| 986 |
+
"grad_norm": 2.243980907997254,
|
| 987 |
+
"learning_rate": 5.447446552494369e-06,
|
| 988 |
+
"loss": 0.5818,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.528003017160098,
|
| 993 |
+
"grad_norm": 2.0846263742726165,
|
| 994 |
+
"learning_rate": 5.381784855472424e-06,
|
| 995 |
+
"loss": 0.5843,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5317744672826702,
|
| 1000 |
+
"grad_norm": 2.1221104917928284,
|
| 1001 |
+
"learning_rate": 5.316056860568318e-06,
|
| 1002 |
+
"loss": 0.5921,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5355459174052423,
|
| 1007 |
+
"grad_norm": 2.12080931648079,
|
| 1008 |
+
"learning_rate": 5.250273981610543e-06,
|
| 1009 |
+
"loss": 0.5782,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5393173675278145,
|
| 1014 |
+
"grad_norm": 2.1035292957689036,
|
| 1015 |
+
"learning_rate": 5.184447641958339e-06,
|
| 1016 |
+
"loss": 0.5822,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.5430888176503865,
|
| 1021 |
+
"grad_norm": 2.06196021971329,
|
| 1022 |
+
"learning_rate": 5.118589272518008e-06,
|
| 1023 |
+
"loss": 0.5728,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.5468602677729587,
|
| 1028 |
+
"grad_norm": 2.055521224880793,
|
| 1029 |
+
"learning_rate": 5.052710309757899e-06,
|
| 1030 |
+
"loss": 0.5879,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.5506317178955308,
|
| 1035 |
+
"grad_norm": 2.1354094712615774,
|
| 1036 |
+
"learning_rate": 4.9868221937224435e-06,
|
| 1037 |
+
"loss": 0.5767,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.554403168018103,
|
| 1042 |
+
"grad_norm": 2.152252412903219,
|
| 1043 |
+
"learning_rate": 4.920936366045564e-06,
|
| 1044 |
+
"loss": 0.5769,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.558174618140675,
|
| 1049 |
+
"grad_norm": 2.2029723753911066,
|
| 1050 |
+
"learning_rate": 4.8550642679637985e-06,
|
| 1051 |
+
"loss": 0.578,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5619460682632472,
|
| 1056 |
+
"grad_norm": 2.129220459700306,
|
| 1057 |
+
"learning_rate": 4.789217338329515e-06,
|
| 1058 |
+
"loss": 0.5599,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.5657175183858193,
|
| 1063 |
+
"grad_norm": 2.3375640799042507,
|
| 1064 |
+
"learning_rate": 4.723407011624509e-06,
|
| 1065 |
+
"loss": 0.5624,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5694889685083915,
|
| 1070 |
+
"grad_norm": 2.086098519936994,
|
| 1071 |
+
"learning_rate": 4.657644715974397e-06,
|
| 1072 |
+
"loss": 0.5594,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.5732604186309636,
|
| 1077 |
+
"grad_norm": 2.0842869926932672,
|
| 1078 |
+
"learning_rate": 4.591941871164075e-06,
|
| 1079 |
+
"loss": 0.5705,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.5770318687535357,
|
| 1084 |
+
"grad_norm": 2.266813054356403,
|
| 1085 |
+
"learning_rate": 4.526309886654659e-06,
|
| 1086 |
+
"loss": 0.554,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.5808033188761078,
|
| 1091 |
+
"grad_norm": 2.1089635718005333,
|
| 1092 |
+
"learning_rate": 4.4607601596021995e-06,
|
| 1093 |
+
"loss": 0.5583,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.58457476899868,
|
| 1098 |
+
"grad_norm": 2.2148767655660984,
|
| 1099 |
+
"learning_rate": 4.395304072878533e-06,
|
| 1100 |
+
"loss": 0.5584,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5883462191212521,
|
| 1105 |
+
"grad_norm": 2.0165033312706164,
|
| 1106 |
+
"learning_rate": 4.329952993094627e-06,
|
| 1107 |
+
"loss": 0.5587,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.5921176692438243,
|
| 1112 |
+
"grad_norm": 2.0613406319368206,
|
| 1113 |
+
"learning_rate": 4.264718268626729e-06,
|
| 1114 |
+
"loss": 0.5549,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5958891193663963,
|
| 1119 |
+
"grad_norm": 2.23703436211283,
|
| 1120 |
+
"learning_rate": 4.199611227645705e-06,
|
| 1121 |
+
"loss": 0.5608,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.5996605694889685,
|
| 1126 |
+
"grad_norm": 2.1229182241921354,
|
| 1127 |
+
"learning_rate": 4.134643176149861e-06,
|
| 1128 |
+
"loss": 0.5423,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.6034320196115406,
|
| 1133 |
+
"grad_norm": 2.16174398622042,
|
| 1134 |
+
"learning_rate": 4.069825396001643e-06,
|
| 1135 |
+
"loss": 0.5496,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.6072034697341128,
|
| 1140 |
+
"grad_norm": 2.128117272383355,
|
| 1141 |
+
"learning_rate": 4.005169142968503e-06,
|
| 1142 |
+
"loss": 0.5519,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.610974919856685,
|
| 1147 |
+
"grad_norm": 2.0175251911767216,
|
| 1148 |
+
"learning_rate": 3.94068564476831e-06,
|
| 1149 |
+
"loss": 0.5475,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.614746369979257,
|
| 1154 |
+
"grad_norm": 2.0746720466907558,
|
| 1155 |
+
"learning_rate": 3.876386099119635e-06,
|
| 1156 |
+
"loss": 0.5461,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.6185178201018292,
|
| 1161 |
+
"grad_norm": 2.130560812479597,
|
| 1162 |
+
"learning_rate": 3.8122816717972287e-06,
|
| 1163 |
+
"loss": 0.5407,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.6222892702244013,
|
| 1168 |
+
"grad_norm": 2.1146940344656375,
|
| 1169 |
+
"learning_rate": 3.7483834946930682e-06,
|
| 1170 |
+
"loss": 0.5504,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.6260607203469735,
|
| 1175 |
+
"grad_norm": 2.174000452771408,
|
| 1176 |
+
"learning_rate": 3.684702663883266e-06,
|
| 1177 |
+
"loss": 0.5497,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.6298321704695455,
|
| 1182 |
+
"grad_norm": 2.173713479741331,
|
| 1183 |
+
"learning_rate": 3.6212502377012164e-06,
|
| 1184 |
+
"loss": 0.5456,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.6336036205921177,
|
| 1189 |
+
"grad_norm": 2.1517414568189275,
|
| 1190 |
+
"learning_rate": 3.558037234817291e-06,
|
| 1191 |
+
"loss": 0.5344,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.6373750707146898,
|
| 1196 |
+
"grad_norm": 2.0901545430594837,
|
| 1197 |
+
"learning_rate": 3.495074632325407e-06,
|
| 1198 |
+
"loss": 0.5357,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.641146520837262,
|
| 1203 |
+
"grad_norm": 2.1721190985604073,
|
| 1204 |
+
"learning_rate": 3.4323733638368427e-06,
|
| 1205 |
+
"loss": 0.5366,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.644917970959834,
|
| 1210 |
+
"grad_norm": 2.1159591065311836,
|
| 1211 |
+
"learning_rate": 3.3699443175815773e-06,
|
| 1212 |
+
"loss": 0.5209,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.6486894210824062,
|
| 1217 |
+
"grad_norm": 2.092825596814058,
|
| 1218 |
+
"learning_rate": 3.3077983345175295e-06,
|
| 1219 |
+
"loss": 0.5357,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6524608712049783,
|
| 1224 |
+
"grad_norm": 2.112302580054298,
|
| 1225 |
+
"learning_rate": 3.2459462064479972e-06,
|
| 1226 |
+
"loss": 0.5399,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.6562323213275505,
|
| 1231 |
+
"grad_norm": 2.118173020663562,
|
| 1232 |
+
"learning_rate": 3.1843986741476285e-06,
|
| 1233 |
+
"loss": 0.5381,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.6600037714501226,
|
| 1238 |
+
"grad_norm": 2.089034293925172,
|
| 1239 |
+
"learning_rate": 3.1231664254972736e-06,
|
| 1240 |
+
"loss": 0.5239,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.6637752215726948,
|
| 1245 |
+
"grad_norm": 2.129522823953631,
|
| 1246 |
+
"learning_rate": 3.0622600936279914e-06,
|
| 1247 |
+
"loss": 0.5461,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.6675466716952668,
|
| 1252 |
+
"grad_norm": 2.0687230926828835,
|
| 1253 |
+
"learning_rate": 3.0016902550745896e-06,
|
| 1254 |
+
"loss": 0.5315,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.671318121817839,
|
| 1259 |
+
"grad_norm": 2.1182888988383146,
|
| 1260 |
+
"learning_rate": 2.941467427938981e-06,
|
| 1261 |
+
"loss": 0.5251,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.6750895719404111,
|
| 1266 |
+
"grad_norm": 2.166050854772299,
|
| 1267 |
+
"learning_rate": 2.881602070063695e-06,
|
| 1268 |
+
"loss": 0.5313,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.6788610220629833,
|
| 1273 |
+
"grad_norm": 2.119609605339128,
|
| 1274 |
+
"learning_rate": 2.822104577215831e-06,
|
| 1275 |
+
"loss": 0.5298,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.6826324721855553,
|
| 1280 |
+
"grad_norm": 2.3123977751286895,
|
| 1281 |
+
"learning_rate": 2.762985281281828e-06,
|
| 1282 |
+
"loss": 0.5162,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.6864039223081275,
|
| 1287 |
+
"grad_norm": 2.147092195126045,
|
| 1288 |
+
"learning_rate": 2.7042544484732958e-06,
|
| 1289 |
+
"loss": 0.5225,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.6901753724306996,
|
| 1294 |
+
"grad_norm": 2.1358012203129064,
|
| 1295 |
+
"learning_rate": 2.6459222775442582e-06,
|
| 1296 |
+
"loss": 0.5083,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.6939468225532718,
|
| 1301 |
+
"grad_norm": 2.086154110222615,
|
| 1302 |
+
"learning_rate": 2.587998898020131e-06,
|
| 1303 |
+
"loss": 0.5163,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.6977182726758439,
|
| 1308 |
+
"grad_norm": 2.1069385491797905,
|
| 1309 |
+
"learning_rate": 2.530494368438683e-06,
|
| 1310 |
+
"loss": 0.5131,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.701489722798416,
|
| 1315 |
+
"grad_norm": 2.0817537626792757,
|
| 1316 |
+
"learning_rate": 2.473418674603363e-06,
|
| 1317 |
+
"loss": 0.5121,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.7052611729209881,
|
| 1322 |
+
"grad_norm": 2.177796705379638,
|
| 1323 |
+
"learning_rate": 2.4167817278492337e-06,
|
| 1324 |
+
"loss": 0.5085,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.7090326230435603,
|
| 1329 |
+
"grad_norm": 2.0422929484726238,
|
| 1330 |
+
"learning_rate": 2.3605933633218396e-06,
|
| 1331 |
+
"loss": 0.5126,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.7128040731661324,
|
| 1336 |
+
"grad_norm": 2.0950516494712907,
|
| 1337 |
+
"learning_rate": 2.304863338269326e-06,
|
| 1338 |
+
"loss": 0.5019,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.7165755232887046,
|
| 1343 |
+
"grad_norm": 2.139347400411628,
|
| 1344 |
+
"learning_rate": 2.249601330348044e-06,
|
| 1345 |
+
"loss": 0.5181,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.7203469734112766,
|
| 1350 |
+
"grad_norm": 2.0787373792617103,
|
| 1351 |
+
"learning_rate": 2.194816935942024e-06,
|
| 1352 |
+
"loss": 0.5095,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.7241184235338488,
|
| 1357 |
+
"grad_norm": 2.0934162351579007,
|
| 1358 |
+
"learning_rate": 2.140519668496529e-06,
|
| 1359 |
+
"loss": 0.5155,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.7278898736564209,
|
| 1364 |
+
"grad_norm": 2.118132732572834,
|
| 1365 |
+
"learning_rate": 2.086718956866024e-06,
|
| 1366 |
+
"loss": 0.4989,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.7316613237789931,
|
| 1371 |
+
"grad_norm": 2.015918946806562,
|
| 1372 |
+
"learning_rate": 2.033424143676825e-06,
|
| 1373 |
+
"loss": 0.4949,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.7354327739015651,
|
| 1378 |
+
"grad_norm": 2.270252257697222,
|
| 1379 |
+
"learning_rate": 1.9806444837047478e-06,
|
| 1380 |
+
"loss": 0.5106,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.7392042240241373,
|
| 1385 |
+
"grad_norm": 2.124764638447049,
|
| 1386 |
+
"learning_rate": 1.928389142267969e-06,
|
| 1387 |
+
"loss": 0.5041,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.7429756741467094,
|
| 1392 |
+
"grad_norm": 2.047944456908165,
|
| 1393 |
+
"learning_rate": 1.8766671936354647e-06,
|
| 1394 |
+
"loss": 0.5092,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.7467471242692816,
|
| 1399 |
+
"grad_norm": 2.0523494780642264,
|
| 1400 |
+
"learning_rate": 1.8254876194512344e-06,
|
| 1401 |
+
"loss": 0.4972,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.7505185743918537,
|
| 1406 |
+
"grad_norm": 2.070102714868304,
|
| 1407 |
+
"learning_rate": 1.7748593071746089e-06,
|
| 1408 |
+
"loss": 0.516,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.7542900245144258,
|
| 1413 |
+
"grad_norm": 2.1874025056879978,
|
| 1414 |
+
"learning_rate": 1.7247910485369417e-06,
|
| 1415 |
+
"loss": 0.504,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.7580614746369979,
|
| 1420 |
+
"grad_norm": 2.1407434998612547,
|
| 1421 |
+
"learning_rate": 1.6752915380148772e-06,
|
| 1422 |
+
"loss": 0.4958,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.7618329247595701,
|
| 1427 |
+
"grad_norm": 2.090141627816375,
|
| 1428 |
+
"learning_rate": 1.626369371320553e-06,
|
| 1429 |
+
"loss": 0.4954,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.7656043748821422,
|
| 1434 |
+
"grad_norm": 2.0433900279296147,
|
| 1435 |
+
"learning_rate": 1.578033043908928e-06,
|
| 1436 |
+
"loss": 0.4994,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.7693758250047144,
|
| 1441 |
+
"grad_norm": 2.3054787165813257,
|
| 1442 |
+
"learning_rate": 1.5302909495025164e-06,
|
| 1443 |
+
"loss": 0.4827,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.7731472751272864,
|
| 1448 |
+
"grad_norm": 2.07475519454638,
|
| 1449 |
+
"learning_rate": 1.4831513786338126e-06,
|
| 1450 |
+
"loss": 0.4983,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.7769187252498586,
|
| 1455 |
+
"grad_norm": 2.1363918208165686,
|
| 1456 |
+
"learning_rate": 1.4366225172056085e-06,
|
| 1457 |
+
"loss": 0.4939,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.7806901753724307,
|
| 1462 |
+
"grad_norm": 2.3053265439992034,
|
| 1463 |
+
"learning_rate": 1.390712445069496e-06,
|
| 1464 |
+
"loss": 0.4903,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.7844616254950029,
|
| 1469 |
+
"grad_norm": 2.0769867503422765,
|
| 1470 |
+
"learning_rate": 1.3454291346227787e-06,
|
| 1471 |
+
"loss": 0.4974,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.7882330756175749,
|
| 1476 |
+
"grad_norm": 2.1092355082866368,
|
| 1477 |
+
"learning_rate": 1.3007804494240478e-06,
|
| 1478 |
+
"loss": 0.4926,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.7920045257401471,
|
| 1483 |
+
"grad_norm": 2.0675161012597387,
|
| 1484 |
+
"learning_rate": 1.2567741428276513e-06,
|
| 1485 |
+
"loss": 0.4876,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.7957759758627192,
|
| 1490 |
+
"grad_norm": 2.1019083361876274,
|
| 1491 |
+
"learning_rate": 1.213417856637311e-06,
|
| 1492 |
+
"loss": 0.4805,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.7995474259852914,
|
| 1497 |
+
"grad_norm": 2.044608822527271,
|
| 1498 |
+
"learning_rate": 1.1707191197791012e-06,
|
| 1499 |
+
"loss": 0.488,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.8033188761078635,
|
| 1504 |
+
"grad_norm": 2.0934676621380035,
|
| 1505 |
+
"learning_rate": 1.128685346994028e-06,
|
| 1506 |
+
"loss": 0.4805,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.8070903262304356,
|
| 1511 |
+
"grad_norm": 2.168897054466299,
|
| 1512 |
+
"learning_rate": 1.0873238375504536e-06,
|
| 1513 |
+
"loss": 0.4966,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.8108617763530077,
|
| 1518 |
+
"grad_norm": 2.0831295415470694,
|
| 1519 |
+
"learning_rate": 1.046641773976544e-06,
|
| 1520 |
+
"loss": 0.4888,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.8146332264755799,
|
| 1525 |
+
"grad_norm": 2.0591460741221517,
|
| 1526 |
+
"learning_rate": 1.0066462208130174e-06,
|
| 1527 |
+
"loss": 0.4808,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.818404676598152,
|
| 1532 |
+
"grad_norm": 2.1413757269828677,
|
| 1533 |
+
"learning_rate": 9.673441233863661e-07,
|
| 1534 |
+
"loss": 0.4846,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.8221761267207242,
|
| 1539 |
+
"grad_norm": 2.0597898524385085,
|
| 1540 |
+
"learning_rate": 9.28742306602779e-07,
|
| 1541 |
+
"loss": 0.4801,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.8259475768432962,
|
| 1546 |
+
"grad_norm": 2.0619533474846072,
|
| 1547 |
+
"learning_rate": 8.908474737629897e-07,
|
| 1548 |
+
"loss": 0.4818,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.8297190269658684,
|
| 1553 |
+
"grad_norm": 2.0785020873929563,
|
| 1554 |
+
"learning_rate": 8.536662053982181e-07,
|
| 1555 |
+
"loss": 0.4796,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.8334904770884405,
|
| 1560 |
+
"grad_norm": 2.0473348850093886,
|
| 1561 |
+
"learning_rate": 8.172049581274571e-07,
|
| 1562 |
+
"loss": 0.4818,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.8372619272110127,
|
| 1567 |
+
"grad_norm": 2.0982171571132793,
|
| 1568 |
+
"learning_rate": 7.814700635362598e-07,
|
| 1569 |
+
"loss": 0.473,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.8410333773335847,
|
| 1574 |
+
"grad_norm": 2.1276785014387114,
|
| 1575 |
+
"learning_rate": 7.464677270772441e-07,
|
| 1576 |
+
"loss": 0.4779,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.8448048274561569,
|
| 1581 |
+
"grad_norm": 2.1127490084192146,
|
| 1582 |
+
"learning_rate": 7.122040269924985e-07,
|
| 1583 |
+
"loss": 0.4741,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.848576277578729,
|
| 1588 |
+
"grad_norm": 2.165957082369384,
|
| 1589 |
+
"learning_rate": 6.786849132580841e-07,
|
| 1590 |
+
"loss": 0.4797,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.8523477277013012,
|
| 1595 |
+
"grad_norm": 2.0518592705151115,
|
| 1596 |
+
"learning_rate": 6.459162065508067e-07,
|
| 1597 |
+
"loss": 0.478,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.8561191778238733,
|
| 1602 |
+
"grad_norm": 2.166898922792351,
|
| 1603 |
+
"learning_rate": 6.139035972374374e-07,
|
| 1604 |
+
"loss": 0.4817,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.8598906279464454,
|
| 1609 |
+
"grad_norm": 2.0882349445622808,
|
| 1610 |
+
"learning_rate": 5.826526443865732e-07,
|
| 1611 |
+
"loss": 0.4774,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.8636620780690175,
|
| 1616 |
+
"grad_norm": 2.0860313130166364,
|
| 1617 |
+
"learning_rate": 5.521687748032805e-07,
|
| 1618 |
+
"loss": 0.4808,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.8674335281915897,
|
| 1623 |
+
"grad_norm": 2.178027486209201,
|
| 1624 |
+
"learning_rate": 5.224572820867324e-07,
|
| 1625 |
+
"loss": 0.4818,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.8712049783141618,
|
| 1630 |
+
"grad_norm": 2.147704424651778,
|
| 1631 |
+
"learning_rate": 4.935233257109456e-07,
|
| 1632 |
+
"loss": 0.4768,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.874976428436734,
|
| 1637 |
+
"grad_norm": 2.070907197952466,
|
| 1638 |
+
"learning_rate": 4.653719301288362e-07,
|
| 1639 |
+
"loss": 0.4876,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.878747878559306,
|
| 1644 |
+
"grad_norm": 2.052204629034235,
|
| 1645 |
+
"learning_rate": 4.380079838997087e-07,
|
| 1646 |
+
"loss": 0.4746,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.8825193286818782,
|
| 1651 |
+
"grad_norm": 2.0244144715714176,
|
| 1652 |
+
"learning_rate": 4.114362388403409e-07,
|
| 1653 |
+
"loss": 0.4804,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.8862907788044503,
|
| 1658 |
+
"grad_norm": 1.997146927651021,
|
| 1659 |
+
"learning_rate": 3.8566130919982567e-07,
|
| 1660 |
+
"loss": 0.4711,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.8900622289270225,
|
| 1665 |
+
"grad_norm": 2.0695766404157574,
|
| 1666 |
+
"learning_rate": 3.606876708582918e-07,
|
| 1667 |
+
"loss": 0.4735,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.8938336790495945,
|
| 1672 |
+
"grad_norm": 2.088709302750632,
|
| 1673 |
+
"learning_rate": 3.3651966054965734e-07,
|
| 1674 |
+
"loss": 0.4811,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.8976051291721667,
|
| 1679 |
+
"grad_norm": 2.101334710434882,
|
| 1680 |
+
"learning_rate": 3.1316147510854434e-07,
|
| 1681 |
+
"loss": 0.4696,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.9013765792947388,
|
| 1686 |
+
"grad_norm": 2.085312151221947,
|
| 1687 |
+
"learning_rate": 2.9061717074148623e-07,
|
| 1688 |
+
"loss": 0.4763,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.905148029417311,
|
| 1693 |
+
"grad_norm": 2.139678602127972,
|
| 1694 |
+
"learning_rate": 2.688906623225629e-07,
|
| 1695 |
+
"loss": 0.4777,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.9089194795398831,
|
| 1700 |
+
"grad_norm": 2.1249347744645086,
|
| 1701 |
+
"learning_rate": 2.479857227135685e-07,
|
| 1702 |
+
"loss": 0.4709,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.9126909296624552,
|
| 1707 |
+
"grad_norm": 2.108323773956457,
|
| 1708 |
+
"learning_rate": 2.2790598210884695e-07,
|
| 1709 |
+
"loss": 0.476,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.9164623797850273,
|
| 1714 |
+
"grad_norm": 2.0522415633167,
|
| 1715 |
+
"learning_rate": 2.0865492740489822e-07,
|
| 1716 |
+
"loss": 0.4663,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.9202338299075995,
|
| 1721 |
+
"grad_norm": 2.015502653559595,
|
| 1722 |
+
"learning_rate": 1.9023590159487237e-07,
|
| 1723 |
+
"loss": 0.4678,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.9240052800301716,
|
| 1728 |
+
"grad_norm": 2.1033783508867647,
|
| 1729 |
+
"learning_rate": 1.7265210318804683e-07,
|
| 1730 |
+
"loss": 0.4735,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.9277767301527438,
|
| 1735 |
+
"grad_norm": 2.0849738071956017,
|
| 1736 |
+
"learning_rate": 1.5590658565440063e-07,
|
| 1737 |
+
"loss": 0.4814,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.9315481802753158,
|
| 1742 |
+
"grad_norm": 2.1194015599967018,
|
| 1743 |
+
"learning_rate": 1.4000225689437263e-07,
|
| 1744 |
+
"loss": 0.4767,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.935319630397888,
|
| 1749 |
+
"grad_norm": 2.155785238683729,
|
| 1750 |
+
"learning_rate": 1.2494187873389186e-07,
|
| 1751 |
+
"loss": 0.4702,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.9390910805204601,
|
| 1756 |
+
"grad_norm": 2.1492733094402428,
|
| 1757 |
+
"learning_rate": 1.107280664447874e-07,
|
| 1758 |
+
"loss": 0.4678,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.9428625306430323,
|
| 1763 |
+
"grad_norm": 2.0766452818387973,
|
| 1764 |
+
"learning_rate": 9.73632882906328e-08,
|
| 1765 |
+
"loss": 0.4716,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.9466339807656043,
|
| 1770 |
+
"grad_norm": 2.0669365458508513,
|
| 1771 |
+
"learning_rate": 8.484986509813165e-08,
|
| 1772 |
+
"loss": 0.4651,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.9504054308881765,
|
| 1777 |
+
"grad_norm": 2.1012750712142587,
|
| 1778 |
+
"learning_rate": 7.318996985409609e-08,
|
| 1779 |
+
"loss": 0.4744,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.9541768810107486,
|
| 1784 |
+
"grad_norm": 2.0562321019527188,
|
| 1785 |
+
"learning_rate": 6.238562732810427e-08,
|
| 1786 |
+
"loss": 0.4626,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.9579483311333208,
|
| 1791 |
+
"grad_norm": 2.2665109481569305,
|
| 1792 |
+
"learning_rate": 5.243871372089482e-08,
|
| 1793 |
+
"loss": 0.4756,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.9617197812558929,
|
| 1798 |
+
"grad_norm": 2.01339402739892,
|
| 1799 |
+
"learning_rate": 4.335095633855868e-08,
|
| 1800 |
+
"loss": 0.4701,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.965491231378465,
|
| 1805 |
+
"grad_norm": 2.0355724329334057,
|
| 1806 |
+
"learning_rate": 3.512393329258734e-08,
|
| 1807 |
+
"loss": 0.4681,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.9692626815010371,
|
| 1812 |
+
"grad_norm": 2.0763696915453793,
|
| 1813 |
+
"learning_rate": 2.77590732258326e-08,
|
| 1814 |
+
"loss": 0.4853,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.9730341316236093,
|
| 1819 |
+
"grad_norm": 2.223079040055733,
|
| 1820 |
+
"learning_rate": 2.1257655064416684e-08,
|
| 1821 |
+
"loss": 0.4755,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.9768055817461814,
|
| 1826 |
+
"grad_norm": 2.0350565583163522,
|
| 1827 |
+
"learning_rate": 1.56208077956449e-08,
|
| 1828 |
+
"loss": 0.4644,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.9805770318687536,
|
| 1833 |
+
"grad_norm": 2.0017384058730805,
|
| 1834 |
+
"learning_rate": 1.0849510271953578e-08,
|
| 1835 |
+
"loss": 0.4785,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.9843484819913256,
|
| 1840 |
+
"grad_norm": 2.0937310226370687,
|
| 1841 |
+
"learning_rate": 6.944591040930481e-09,
|
| 1842 |
+
"loss": 0.4697,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.9881199321138978,
|
| 1847 |
+
"grad_norm": 2.0794638687177938,
|
| 1848 |
+
"learning_rate": 3.9067282014354504e-09,
|
| 1849 |
+
"loss": 0.4707,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.9918913822364699,
|
| 1854 |
+
"grad_norm": 2.0578690852807098,
|
| 1855 |
+
"learning_rate": 1.7364492858451542e-09,
|
| 1856 |
+
"loss": 0.468,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.9956628323590421,
|
| 1861 |
+
"grad_norm": 2.1418033123461093,
|
| 1862 |
+
"learning_rate": 4.341311684485927e-10,
|
| 1863 |
+
"loss": 0.4788,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.9994342824816141,
|
| 1868 |
+
"grad_norm": 2.041628364481552,
|
| 1869 |
+
"learning_rate": 0.0,
|
| 1870 |
+
"loss": 0.4741,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.9994342824816141,
|
| 1875 |
+
"eval_loss": 0.47078943252563477,
|
| 1876 |
+
"eval_runtime": 251.4407,
|
| 1877 |
+
"eval_samples_per_second": 3.094,
|
| 1878 |
+
"eval_steps_per_second": 0.776,
|
| 1879 |
+
"step": 1325
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.9994342824816141,
|
| 1883 |
+
"step": 1325,
|
| 1884 |
+
"total_flos": 277375698862080.0,
|
| 1885 |
+
"train_loss": 0.6321603435840247,
|
| 1886 |
+
"train_runtime": 30135.1647,
|
| 1887 |
+
"train_samples_per_second": 1.408,
|
| 1888 |
+
"train_steps_per_second": 0.044
|
| 1889 |
+
}
|
| 1890 |
+
],
|
| 1891 |
+
"logging_steps": 5,
|
| 1892 |
+
"max_steps": 1325,
|
| 1893 |
+
"num_input_tokens_seen": 0,
|
| 1894 |
+
"num_train_epochs": 1,
|
| 1895 |
+
"save_steps": 100,
|
| 1896 |
+
"stateful_callbacks": {
|
| 1897 |
+
"TrainerControl": {
|
| 1898 |
+
"args": {
|
| 1899 |
+
"should_epoch_stop": false,
|
| 1900 |
+
"should_evaluate": false,
|
| 1901 |
+
"should_log": false,
|
| 1902 |
+
"should_save": true,
|
| 1903 |
+
"should_training_stop": true
|
| 1904 |
+
},
|
| 1905 |
+
"attributes": {}
|
| 1906 |
+
}
|
| 1907 |
+
},
|
| 1908 |
+
"total_flos": 277375698862080.0,
|
| 1909 |
+
"train_batch_size": 2,
|
| 1910 |
+
"trial_name": null,
|
| 1911 |
+
"trial_params": null
|
| 1912 |
+
}
|