Commit
•
2cee902
1
Parent(s):
6380931
Model save
Browse files- README.md +69 -0
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +2607 -0
README.md
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
|
| 3 |
+
library_name: peft
|
| 4 |
+
license: llama3.1
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: barc-llama3.1-8b-instruct-lora64-induction-gpt4-desc-llama-20k_lr2e-4_epoch3
|
| 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 |
+
# barc-llama3.1-8b-instruct-lora64-induction-gpt4-desc-llama-20k_lr2e-4_epoch3
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.1384
|
| 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: 8
|
| 42 |
+
- eval_batch_size: 4
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 8
|
| 46 |
+
- gradient_accumulation_steps: 2
|
| 47 |
+
- total_train_batch_size: 128
|
| 48 |
+
- total_eval_batch_size: 32
|
| 49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 50 |
+
- lr_scheduler_type: cosine
|
| 51 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 52 |
+
- num_epochs: 3
|
| 53 |
+
|
| 54 |
+
### Training results
|
| 55 |
+
|
| 56 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 57 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 58 |
+
| 0.1393 | 0.9959 | 121 | 0.1582 |
|
| 59 |
+
| 0.1282 | 2.0 | 243 | 0.1401 |
|
| 60 |
+
| 0.1135 | 2.9877 | 363 | 0.1384 |
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
### Framework versions
|
| 64 |
+
|
| 65 |
+
- PEFT 0.13.0
|
| 66 |
+
- Transformers 4.45.0.dev0
|
| 67 |
+
- Pytorch 2.4.1+cu121
|
| 68 |
+
- Datasets 3.0.0
|
| 69 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9876543209876543,
|
| 3 |
+
"total_flos": 9.618537139981517e+16,
|
| 4 |
+
"train_loss": 0.21143763487742953,
|
| 5 |
+
"train_runtime": 3172.33,
|
| 6 |
+
"train_samples": 15511,
|
| 7 |
+
"train_samples_per_second": 14.668,
|
| 8 |
+
"train_steps_per_second": 0.114
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9876543209876543,
|
| 3 |
+
"total_flos": 9.618537139981517e+16,
|
| 4 |
+
"train_loss": 0.21143763487742953,
|
| 5 |
+
"train_runtime": 3172.33,
|
| 6 |
+
"train_samples": 15511,
|
| 7 |
+
"train_samples_per_second": 14.668,
|
| 8 |
+
"train_steps_per_second": 0.114
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2607 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9876543209876543,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 363,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.00823045267489712,
|
| 13 |
+
"grad_norm": 1.4866708861109026,
|
| 14 |
+
"learning_rate": 5.405405405405406e-06,
|
| 15 |
+
"loss": 1.7565,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.01646090534979424,
|
| 20 |
+
"grad_norm": 1.4564242960595204,
|
| 21 |
+
"learning_rate": 1.0810810810810812e-05,
|
| 22 |
+
"loss": 1.7428,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.024691358024691357,
|
| 27 |
+
"grad_norm": 1.5029588271629886,
|
| 28 |
+
"learning_rate": 1.6216216216216218e-05,
|
| 29 |
+
"loss": 1.7399,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.03292181069958848,
|
| 34 |
+
"grad_norm": 1.4275061812775922,
|
| 35 |
+
"learning_rate": 2.1621621621621624e-05,
|
| 36 |
+
"loss": 1.7083,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0411522633744856,
|
| 41 |
+
"grad_norm": 1.3211892837590813,
|
| 42 |
+
"learning_rate": 2.702702702702703e-05,
|
| 43 |
+
"loss": 1.6472,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.04938271604938271,
|
| 48 |
+
"grad_norm": 1.2954095745224536,
|
| 49 |
+
"learning_rate": 3.2432432432432436e-05,
|
| 50 |
+
"loss": 1.5935,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.05761316872427984,
|
| 55 |
+
"grad_norm": 0.9115980406962145,
|
| 56 |
+
"learning_rate": 3.783783783783784e-05,
|
| 57 |
+
"loss": 1.4303,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.06584362139917696,
|
| 62 |
+
"grad_norm": 0.7713124020101795,
|
| 63 |
+
"learning_rate": 4.324324324324325e-05,
|
| 64 |
+
"loss": 1.3661,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.07407407407407407,
|
| 69 |
+
"grad_norm": 0.7920851957124333,
|
| 70 |
+
"learning_rate": 4.8648648648648654e-05,
|
| 71 |
+
"loss": 1.3093,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0823045267489712,
|
| 76 |
+
"grad_norm": 0.8823079310051533,
|
| 77 |
+
"learning_rate": 5.405405405405406e-05,
|
| 78 |
+
"loss": 1.1587,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.09053497942386832,
|
| 83 |
+
"grad_norm": 0.8557863491129706,
|
| 84 |
+
"learning_rate": 5.9459459459459466e-05,
|
| 85 |
+
"loss": 1.0485,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.09876543209876543,
|
| 90 |
+
"grad_norm": 0.8802985405555526,
|
| 91 |
+
"learning_rate": 6.486486486486487e-05,
|
| 92 |
+
"loss": 0.9429,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.10699588477366255,
|
| 97 |
+
"grad_norm": 0.8436384566917278,
|
| 98 |
+
"learning_rate": 7.027027027027028e-05,
|
| 99 |
+
"loss": 0.8218,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.11522633744855967,
|
| 104 |
+
"grad_norm": 0.6676398141327192,
|
| 105 |
+
"learning_rate": 7.567567567567568e-05,
|
| 106 |
+
"loss": 0.7054,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.12345679012345678,
|
| 111 |
+
"grad_norm": 0.6471842089986185,
|
| 112 |
+
"learning_rate": 8.108108108108109e-05,
|
| 113 |
+
"loss": 0.6758,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.13168724279835392,
|
| 118 |
+
"grad_norm": 0.4142479004737112,
|
| 119 |
+
"learning_rate": 8.64864864864865e-05,
|
| 120 |
+
"loss": 0.6222,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.13991769547325103,
|
| 125 |
+
"grad_norm": 0.42850243250289555,
|
| 126 |
+
"learning_rate": 9.18918918918919e-05,
|
| 127 |
+
"loss": 0.5962,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.14814814814814814,
|
| 132 |
+
"grad_norm": 0.5568208902875614,
|
| 133 |
+
"learning_rate": 9.729729729729731e-05,
|
| 134 |
+
"loss": 0.5984,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.15637860082304528,
|
| 139 |
+
"grad_norm": 0.5540687236420326,
|
| 140 |
+
"learning_rate": 0.0001027027027027027,
|
| 141 |
+
"loss": 0.4938,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.1646090534979424,
|
| 146 |
+
"grad_norm": 0.33332950971206693,
|
| 147 |
+
"learning_rate": 0.00010810810810810812,
|
| 148 |
+
"loss": 0.5161,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.1728395061728395,
|
| 153 |
+
"grad_norm": 0.2653066423031725,
|
| 154 |
+
"learning_rate": 0.00011351351351351351,
|
| 155 |
+
"loss": 0.5136,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.18106995884773663,
|
| 160 |
+
"grad_norm": 0.2272101382108708,
|
| 161 |
+
"learning_rate": 0.00011891891891891893,
|
| 162 |
+
"loss": 0.4873,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.18930041152263374,
|
| 167 |
+
"grad_norm": 0.24006404965358885,
|
| 168 |
+
"learning_rate": 0.00012432432432432433,
|
| 169 |
+
"loss": 0.4844,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.19753086419753085,
|
| 174 |
+
"grad_norm": 0.20608476733616163,
|
| 175 |
+
"learning_rate": 0.00012972972972972974,
|
| 176 |
+
"loss": 0.4821,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.205761316872428,
|
| 181 |
+
"grad_norm": 0.16150407452920948,
|
| 182 |
+
"learning_rate": 0.00013513513513513514,
|
| 183 |
+
"loss": 0.4167,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.2139917695473251,
|
| 188 |
+
"grad_norm": 0.1538751616515209,
|
| 189 |
+
"learning_rate": 0.00014054054054054056,
|
| 190 |
+
"loss": 0.446,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.2222222222222222,
|
| 195 |
+
"grad_norm": 0.16472183357224798,
|
| 196 |
+
"learning_rate": 0.00014594594594594595,
|
| 197 |
+
"loss": 0.4142,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.23045267489711935,
|
| 202 |
+
"grad_norm": 0.1552702492617925,
|
| 203 |
+
"learning_rate": 0.00015135135135135137,
|
| 204 |
+
"loss": 0.4909,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.23868312757201646,
|
| 209 |
+
"grad_norm": 0.14727903417905372,
|
| 210 |
+
"learning_rate": 0.00015675675675675676,
|
| 211 |
+
"loss": 0.4065,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.24691358024691357,
|
| 216 |
+
"grad_norm": 0.14784235244019464,
|
| 217 |
+
"learning_rate": 0.00016216216216216218,
|
| 218 |
+
"loss": 0.3886,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.2551440329218107,
|
| 223 |
+
"grad_norm": 0.1450616035553425,
|
| 224 |
+
"learning_rate": 0.00016756756756756757,
|
| 225 |
+
"loss": 0.3902,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.26337448559670784,
|
| 230 |
+
"grad_norm": 0.153342097650188,
|
| 231 |
+
"learning_rate": 0.000172972972972973,
|
| 232 |
+
"loss": 0.3888,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.2716049382716049,
|
| 237 |
+
"grad_norm": 0.15900713930790533,
|
| 238 |
+
"learning_rate": 0.00017837837837837839,
|
| 239 |
+
"loss": 0.3524,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.27983539094650206,
|
| 244 |
+
"grad_norm": 0.1546499892536795,
|
| 245 |
+
"learning_rate": 0.0001837837837837838,
|
| 246 |
+
"loss": 0.3648,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.2880658436213992,
|
| 251 |
+
"grad_norm": 0.14320964717963003,
|
| 252 |
+
"learning_rate": 0.0001891891891891892,
|
| 253 |
+
"loss": 0.3658,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.2962962962962963,
|
| 258 |
+
"grad_norm": 0.13103727806567708,
|
| 259 |
+
"learning_rate": 0.00019459459459459462,
|
| 260 |
+
"loss": 0.3496,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.3045267489711934,
|
| 265 |
+
"grad_norm": 0.13308383037195082,
|
| 266 |
+
"learning_rate": 0.0002,
|
| 267 |
+
"loss": 0.3572,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.31275720164609055,
|
| 272 |
+
"grad_norm": 0.1387871382009263,
|
| 273 |
+
"learning_rate": 0.00019999535665248002,
|
| 274 |
+
"loss": 0.3214,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.32098765432098764,
|
| 279 |
+
"grad_norm": 0.14270131722019058,
|
| 280 |
+
"learning_rate": 0.0001999814270411335,
|
| 281 |
+
"loss": 0.3445,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.3292181069958848,
|
| 286 |
+
"grad_norm": 0.1545337969749864,
|
| 287 |
+
"learning_rate": 0.000199958212459561,
|
| 288 |
+
"loss": 0.32,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.3374485596707819,
|
| 293 |
+
"grad_norm": 0.13802182056866283,
|
| 294 |
+
"learning_rate": 0.00019992571506363,
|
| 295 |
+
"loss": 0.3449,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.345679012345679,
|
| 300 |
+
"grad_norm": 0.14162173979281106,
|
| 301 |
+
"learning_rate": 0.00019988393787127441,
|
| 302 |
+
"loss": 0.3262,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.35390946502057613,
|
| 307 |
+
"grad_norm": 0.13594846523403772,
|
| 308 |
+
"learning_rate": 0.0001998328847622148,
|
| 309 |
+
"loss": 0.2958,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.36213991769547327,
|
| 314 |
+
"grad_norm": 0.12596629667831052,
|
| 315 |
+
"learning_rate": 0.00019977256047759765,
|
| 316 |
+
"loss": 0.286,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.37037037037037035,
|
| 321 |
+
"grad_norm": 0.13835455454793713,
|
| 322 |
+
"learning_rate": 0.00019970297061955533,
|
| 323 |
+
"loss": 0.2878,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.3786008230452675,
|
| 328 |
+
"grad_norm": 0.14644104388608956,
|
| 329 |
+
"learning_rate": 0.00019962412165068573,
|
| 330 |
+
"loss": 0.2952,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.3868312757201646,
|
| 335 |
+
"grad_norm": 0.1361181054099992,
|
| 336 |
+
"learning_rate": 0.00019953602089345217,
|
| 337 |
+
"loss": 0.267,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.3950617283950617,
|
| 342 |
+
"grad_norm": 0.1545760777310844,
|
| 343 |
+
"learning_rate": 0.0001994386765295032,
|
| 344 |
+
"loss": 0.2823,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.40329218106995884,
|
| 349 |
+
"grad_norm": 0.16048561280797768,
|
| 350 |
+
"learning_rate": 0.00019933209759891317,
|
| 351 |
+
"loss": 0.2598,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.411522633744856,
|
| 356 |
+
"grad_norm": 0.14585143788717048,
|
| 357 |
+
"learning_rate": 0.00019921629399934223,
|
| 358 |
+
"loss": 0.2834,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.41975308641975306,
|
| 363 |
+
"grad_norm": 0.15590757372746555,
|
| 364 |
+
"learning_rate": 0.00019909127648511755,
|
| 365 |
+
"loss": 0.2619,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.4279835390946502,
|
| 370 |
+
"grad_norm": 0.142205678471833,
|
| 371 |
+
"learning_rate": 0.0001989570566662345,
|
| 372 |
+
"loss": 0.2477,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.43621399176954734,
|
| 377 |
+
"grad_norm": 0.14334288636686987,
|
| 378 |
+
"learning_rate": 0.00019881364700727823,
|
| 379 |
+
"loss": 0.2958,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.4444444444444444,
|
| 384 |
+
"grad_norm": 0.1584716514858334,
|
| 385 |
+
"learning_rate": 0.0001986610608262665,
|
| 386 |
+
"loss": 0.2708,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.45267489711934156,
|
| 391 |
+
"grad_norm": 0.15229425624208448,
|
| 392 |
+
"learning_rate": 0.00019849931229341258,
|
| 393 |
+
"loss": 0.2776,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.4609053497942387,
|
| 398 |
+
"grad_norm": 0.1497514774482752,
|
| 399 |
+
"learning_rate": 0.00019832841642980945,
|
| 400 |
+
"loss": 0.2325,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.4691358024691358,
|
| 405 |
+
"grad_norm": 0.15302882739571017,
|
| 406 |
+
"learning_rate": 0.00019814838910603481,
|
| 407 |
+
"loss": 0.2755,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.4773662551440329,
|
| 412 |
+
"grad_norm": 0.20080266795074445,
|
| 413 |
+
"learning_rate": 0.00019795924704067721,
|
| 414 |
+
"loss": 0.2421,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.48559670781893005,
|
| 419 |
+
"grad_norm": 0.15063814452987448,
|
| 420 |
+
"learning_rate": 0.00019776100779878345,
|
| 421 |
+
"loss": 0.2152,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.49382716049382713,
|
| 426 |
+
"grad_norm": 0.1424158898404743,
|
| 427 |
+
"learning_rate": 0.00019755368979022732,
|
| 428 |
+
"loss": 0.2424,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.5020576131687243,
|
| 433 |
+
"grad_norm": 0.15708199535695755,
|
| 434 |
+
"learning_rate": 0.00019733731226800015,
|
| 435 |
+
"loss": 0.2439,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.5102880658436214,
|
| 440 |
+
"grad_norm": 0.1469505519285418,
|
| 441 |
+
"learning_rate": 0.00019711189532642243,
|
| 442 |
+
"loss": 0.2174,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.5185185185185185,
|
| 447 |
+
"grad_norm": 0.15443629035666892,
|
| 448 |
+
"learning_rate": 0.00019687745989927823,
|
| 449 |
+
"loss": 0.2201,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.5267489711934157,
|
| 454 |
+
"grad_norm": 0.16091185410714737,
|
| 455 |
+
"learning_rate": 0.00019663402775787066,
|
| 456 |
+
"loss": 0.2153,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.5349794238683128,
|
| 461 |
+
"grad_norm": 0.16812247962269722,
|
| 462 |
+
"learning_rate": 0.00019638162150900027,
|
| 463 |
+
"loss": 0.2245,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.5432098765432098,
|
| 468 |
+
"grad_norm": 0.15476006323853086,
|
| 469 |
+
"learning_rate": 0.00019612026459286578,
|
| 470 |
+
"loss": 0.2168,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.551440329218107,
|
| 475 |
+
"grad_norm": 0.14319948869013133,
|
| 476 |
+
"learning_rate": 0.00019584998128088684,
|
| 477 |
+
"loss": 0.2102,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.5596707818930041,
|
| 482 |
+
"grad_norm": 0.1416595781349322,
|
| 483 |
+
"learning_rate": 0.0001955707966734505,
|
| 484 |
+
"loss": 0.2109,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.5679012345679012,
|
| 489 |
+
"grad_norm": 0.15627936343326146,
|
| 490 |
+
"learning_rate": 0.00019528273669757972,
|
| 491 |
+
"loss": 0.221,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.5761316872427984,
|
| 496 |
+
"grad_norm": 0.1355242418955904,
|
| 497 |
+
"learning_rate": 0.0001949858281045261,
|
| 498 |
+
"loss": 0.1934,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.5843621399176955,
|
| 503 |
+
"grad_norm": 0.15837694932534985,
|
| 504 |
+
"learning_rate": 0.00019468009846728513,
|
| 505 |
+
"loss": 0.2106,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.5925925925925926,
|
| 510 |
+
"grad_norm": 0.14575785598463487,
|
| 511 |
+
"learning_rate": 0.00019436557617803595,
|
| 512 |
+
"loss": 0.1958,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.6008230452674898,
|
| 517 |
+
"grad_norm": 0.15004026647926874,
|
| 518 |
+
"learning_rate": 0.00019404229044550433,
|
| 519 |
+
"loss": 0.2111,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.6090534979423868,
|
| 524 |
+
"grad_norm": 0.1531758472218286,
|
| 525 |
+
"learning_rate": 0.00019371027129225042,
|
| 526 |
+
"loss": 0.1796,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.6172839506172839,
|
| 531 |
+
"grad_norm": 0.14522275863691966,
|
| 532 |
+
"learning_rate": 0.0001933695495518804,
|
| 533 |
+
"loss": 0.1879,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.6255144032921811,
|
| 538 |
+
"grad_norm": 0.14778202822977027,
|
| 539 |
+
"learning_rate": 0.00019302015686618326,
|
| 540 |
+
"loss": 0.1783,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.6337448559670782,
|
| 545 |
+
"grad_norm": 0.14202535229938665,
|
| 546 |
+
"learning_rate": 0.0001926621256821922,
|
| 547 |
+
"loss": 0.1672,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.6419753086419753,
|
| 552 |
+
"grad_norm": 0.15532295497905474,
|
| 553 |
+
"learning_rate": 0.00019229548924917146,
|
| 554 |
+
"loss": 0.1894,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.6502057613168725,
|
| 559 |
+
"grad_norm": 0.1298933712297695,
|
| 560 |
+
"learning_rate": 0.00019192028161552847,
|
| 561 |
+
"loss": 0.1626,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.6584362139917695,
|
| 566 |
+
"grad_norm": 0.15486589167638415,
|
| 567 |
+
"learning_rate": 0.0001915365376256519,
|
| 568 |
+
"loss": 0.1829,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.6666666666666666,
|
| 573 |
+
"grad_norm": 0.13198994845547862,
|
| 574 |
+
"learning_rate": 0.00019114429291667583,
|
| 575 |
+
"loss": 0.1827,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.6748971193415638,
|
| 580 |
+
"grad_norm": 0.14486854526136775,
|
| 581 |
+
"learning_rate": 0.00019074358391517023,
|
| 582 |
+
"loss": 0.1711,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.6831275720164609,
|
| 587 |
+
"grad_norm": 0.18136392511258384,
|
| 588 |
+
"learning_rate": 0.00019033444783375804,
|
| 589 |
+
"loss": 0.1852,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.691358024691358,
|
| 594 |
+
"grad_norm": 0.12835927358561786,
|
| 595 |
+
"learning_rate": 0.00018991692266765947,
|
| 596 |
+
"loss": 0.1874,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.6995884773662552,
|
| 601 |
+
"grad_norm": 0.15923203558527596,
|
| 602 |
+
"learning_rate": 0.00018949104719116332,
|
| 603 |
+
"loss": 0.1754,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.7078189300411523,
|
| 608 |
+
"grad_norm": 0.1524830096667991,
|
| 609 |
+
"learning_rate": 0.00018905686095402647,
|
| 610 |
+
"loss": 0.1772,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.7160493827160493,
|
| 615 |
+
"grad_norm": 0.1458630884597557,
|
| 616 |
+
"learning_rate": 0.0001886144042778006,
|
| 617 |
+
"loss": 0.1884,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.7242798353909465,
|
| 622 |
+
"grad_norm": 0.14812531898730455,
|
| 623 |
+
"learning_rate": 0.00018816371825208789,
|
| 624 |
+
"loss": 0.1694,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.7325102880658436,
|
| 629 |
+
"grad_norm": 0.1371838995231899,
|
| 630 |
+
"learning_rate": 0.0001877048447307252,
|
| 631 |
+
"loss": 0.175,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.7407407407407407,
|
| 636 |
+
"grad_norm": 0.12597672532717027,
|
| 637 |
+
"learning_rate": 0.00018723782632789701,
|
| 638 |
+
"loss": 0.1663,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.7489711934156379,
|
| 643 |
+
"grad_norm": 0.13653897293999862,
|
| 644 |
+
"learning_rate": 0.00018676270641417822,
|
| 645 |
+
"loss": 0.1902,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.757201646090535,
|
| 650 |
+
"grad_norm": 0.11918081783294039,
|
| 651 |
+
"learning_rate": 0.0001862795291125063,
|
| 652 |
+
"loss": 0.1662,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.7654320987654321,
|
| 657 |
+
"grad_norm": 0.1333312758081247,
|
| 658 |
+
"learning_rate": 0.0001857883392940837,
|
| 659 |
+
"loss": 0.199,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.7736625514403292,
|
| 664 |
+
"grad_norm": 0.12930618535907987,
|
| 665 |
+
"learning_rate": 0.000185289182574211,
|
| 666 |
+
"loss": 0.1697,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.7818930041152263,
|
| 671 |
+
"grad_norm": 0.1336640654310965,
|
| 672 |
+
"learning_rate": 0.0001847821053080505,
|
| 673 |
+
"loss": 0.1852,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.7901234567901234,
|
| 678 |
+
"grad_norm": 0.14974272672832492,
|
| 679 |
+
"learning_rate": 0.00018426715458632153,
|
| 680 |
+
"loss": 0.1819,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.7983539094650206,
|
| 685 |
+
"grad_norm": 0.11738767625285426,
|
| 686 |
+
"learning_rate": 0.00018374437823092724,
|
| 687 |
+
"loss": 0.1628,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.8065843621399177,
|
| 692 |
+
"grad_norm": 0.1126501945549117,
|
| 693 |
+
"learning_rate": 0.00018321382479051347,
|
| 694 |
+
"loss": 0.1574,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.8148148148148148,
|
| 699 |
+
"grad_norm": 0.14716070780057008,
|
| 700 |
+
"learning_rate": 0.00018267554353596025,
|
| 701 |
+
"loss": 0.1671,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.823045267489712,
|
| 706 |
+
"grad_norm": 0.14236534823623162,
|
| 707 |
+
"learning_rate": 0.0001821295844558062,
|
| 708 |
+
"loss": 0.179,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.831275720164609,
|
| 713 |
+
"grad_norm": 0.14819744720356537,
|
| 714 |
+
"learning_rate": 0.0001815759982516061,
|
| 715 |
+
"loss": 0.1765,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.8395061728395061,
|
| 720 |
+
"grad_norm": 0.1359773653210936,
|
| 721 |
+
"learning_rate": 0.00018101483633322255,
|
| 722 |
+
"loss": 0.1736,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.8477366255144033,
|
| 727 |
+
"grad_norm": 0.12220336917197815,
|
| 728 |
+
"learning_rate": 0.00018044615081405153,
|
| 729 |
+
"loss": 0.1559,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.8559670781893004,
|
| 734 |
+
"grad_norm": 0.12734749296356374,
|
| 735 |
+
"learning_rate": 0.00017986999450618295,
|
| 736 |
+
"loss": 0.1598,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.8641975308641975,
|
| 741 |
+
"grad_norm": 0.13635727971275893,
|
| 742 |
+
"learning_rate": 0.00017928642091549613,
|
| 743 |
+
"loss": 0.1716,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.8724279835390947,
|
| 748 |
+
"grad_norm": 0.1256041209800328,
|
| 749 |
+
"learning_rate": 0.00017869548423669077,
|
| 750 |
+
"loss": 0.1694,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.8806584362139918,
|
| 755 |
+
"grad_norm": 0.156735390007985,
|
| 756 |
+
"learning_rate": 0.00017809723934825405,
|
| 757 |
+
"loss": 0.1711,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.8888888888888888,
|
| 762 |
+
"grad_norm": 0.13613462557969605,
|
| 763 |
+
"learning_rate": 0.00017749174180736442,
|
| 764 |
+
"loss": 0.1575,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.897119341563786,
|
| 769 |
+
"grad_norm": 0.12136550882477753,
|
| 770 |
+
"learning_rate": 0.00017687904784473188,
|
| 771 |
+
"loss": 0.1541,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.9053497942386831,
|
| 776 |
+
"grad_norm": 0.11805606680499708,
|
| 777 |
+
"learning_rate": 0.00017625921435937637,
|
| 778 |
+
"loss": 0.153,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.9135802469135802,
|
| 783 |
+
"grad_norm": 0.12306976139021918,
|
| 784 |
+
"learning_rate": 0.00017563229891334338,
|
| 785 |
+
"loss": 0.1723,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.9218106995884774,
|
| 790 |
+
"grad_norm": 0.12639467162562257,
|
| 791 |
+
"learning_rate": 0.00017499835972635856,
|
| 792 |
+
"loss": 0.1637,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.9300411522633745,
|
| 797 |
+
"grad_norm": 0.11341325633796959,
|
| 798 |
+
"learning_rate": 0.00017435745567042095,
|
| 799 |
+
"loss": 0.1471,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.9382716049382716,
|
| 804 |
+
"grad_norm": 0.12238460464162876,
|
| 805 |
+
"learning_rate": 0.00017370964626433567,
|
| 806 |
+
"loss": 0.1682,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.9465020576131687,
|
| 811 |
+
"grad_norm": 0.12574079950832473,
|
| 812 |
+
"learning_rate": 0.0001730549916681868,
|
| 813 |
+
"loss": 0.1493,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.9547325102880658,
|
| 818 |
+
"grad_norm": 0.13251823016582745,
|
| 819 |
+
"learning_rate": 0.00017239355267775018,
|
| 820 |
+
"loss": 0.1649,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.9629629629629629,
|
| 825 |
+
"grad_norm": 0.11699891536949006,
|
| 826 |
+
"learning_rate": 0.0001717253907188477,
|
| 827 |
+
"loss": 0.1628,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.9711934156378601,
|
| 832 |
+
"grad_norm": 0.11316448680114288,
|
| 833 |
+
"learning_rate": 0.00017105056784164294,
|
| 834 |
+
"loss": 0.1434,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.9794238683127572,
|
| 839 |
+
"grad_norm": 0.11531932007410475,
|
| 840 |
+
"learning_rate": 0.00017036914671487852,
|
| 841 |
+
"loss": 0.1565,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.9876543209876543,
|
| 846 |
+
"grad_norm": 0.11423778544173263,
|
| 847 |
+
"learning_rate": 0.00016968119062005642,
|
| 848 |
+
"loss": 0.1481,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.9958847736625515,
|
| 853 |
+
"grad_norm": 0.10808259099105172,
|
| 854 |
+
"learning_rate": 0.00016898676344556118,
|
| 855 |
+
"loss": 0.1393,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.9958847736625515,
|
| 860 |
+
"eval_loss": 0.1582392454147339,
|
| 861 |
+
"eval_runtime": 24.3633,
|
| 862 |
+
"eval_samples_per_second": 33.534,
|
| 863 |
+
"eval_steps_per_second": 1.067,
|
| 864 |
+
"step": 121
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.0041152263374487,
|
| 868 |
+
"grad_norm": 0.12179637517786698,
|
| 869 |
+
"learning_rate": 0.00016828592968072678,
|
| 870 |
+
"loss": 0.1367,
|
| 871 |
+
"step": 122
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.0123456790123457,
|
| 875 |
+
"grad_norm": 0.12278954055202473,
|
| 876 |
+
"learning_rate": 0.00016757875440984768,
|
| 877 |
+
"loss": 0.1352,
|
| 878 |
+
"step": 123
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.0205761316872428,
|
| 882 |
+
"grad_norm": 0.11262318129903952,
|
| 883 |
+
"learning_rate": 0.0001668653033061347,
|
| 884 |
+
"loss": 0.1319,
|
| 885 |
+
"step": 124
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.02880658436214,
|
| 889 |
+
"grad_norm": 0.10685919423230769,
|
| 890 |
+
"learning_rate": 0.00016614564262561608,
|
| 891 |
+
"loss": 0.1483,
|
| 892 |
+
"step": 125
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.037037037037037,
|
| 896 |
+
"grad_norm": 0.12688594869999706,
|
| 897 |
+
"learning_rate": 0.0001654198392009846,
|
| 898 |
+
"loss": 0.1345,
|
| 899 |
+
"step": 126
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.045267489711934,
|
| 903 |
+
"grad_norm": 0.11380792656533252,
|
| 904 |
+
"learning_rate": 0.0001646879604353908,
|
| 905 |
+
"loss": 0.1435,
|
| 906 |
+
"step": 127
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.0534979423868314,
|
| 910 |
+
"grad_norm": 0.11336135523754398,
|
| 911 |
+
"learning_rate": 0.00016395007429618382,
|
| 912 |
+
"loss": 0.1496,
|
| 913 |
+
"step": 128
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.0617283950617284,
|
| 917 |
+
"grad_norm": 0.11068219074387899,
|
| 918 |
+
"learning_rate": 0.00016320624930859904,
|
| 919 |
+
"loss": 0.1412,
|
| 920 |
+
"step": 129
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.0699588477366255,
|
| 924 |
+
"grad_norm": 0.10745622525472565,
|
| 925 |
+
"learning_rate": 0.00016245655454939474,
|
| 926 |
+
"loss": 0.1294,
|
| 927 |
+
"step": 130
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.0781893004115226,
|
| 931 |
+
"grad_norm": 0.10201633324761311,
|
| 932 |
+
"learning_rate": 0.00016170105964043695,
|
| 933 |
+
"loss": 0.1443,
|
| 934 |
+
"step": 131
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.0864197530864197,
|
| 938 |
+
"grad_norm": 0.1137314706455847,
|
| 939 |
+
"learning_rate": 0.0001609398347422339,
|
| 940 |
+
"loss": 0.1389,
|
| 941 |
+
"step": 132
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.0946502057613168,
|
| 945 |
+
"grad_norm": 0.11204326746406651,
|
| 946 |
+
"learning_rate": 0.00016017295054742046,
|
| 947 |
+
"loss": 0.1422,
|
| 948 |
+
"step": 133
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.102880658436214,
|
| 952 |
+
"grad_norm": 0.10142584272145459,
|
| 953 |
+
"learning_rate": 0.00015940047827419303,
|
| 954 |
+
"loss": 0.1301,
|
| 955 |
+
"step": 134
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.1111111111111112,
|
| 959 |
+
"grad_norm": 0.10382986472663408,
|
| 960 |
+
"learning_rate": 0.00015862248965969604,
|
| 961 |
+
"loss": 0.1388,
|
| 962 |
+
"step": 135
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.1193415637860082,
|
| 966 |
+
"grad_norm": 0.11472916493277631,
|
| 967 |
+
"learning_rate": 0.00015783905695335946,
|
| 968 |
+
"loss": 0.1406,
|
| 969 |
+
"step": 136
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.1275720164609053,
|
| 973 |
+
"grad_norm": 0.11056711973862805,
|
| 974 |
+
"learning_rate": 0.0001570502529101896,
|
| 975 |
+
"loss": 0.1295,
|
| 976 |
+
"step": 137
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.1358024691358024,
|
| 980 |
+
"grad_norm": 0.11881161089888494,
|
| 981 |
+
"learning_rate": 0.00015625615078401244,
|
| 982 |
+
"loss": 0.1491,
|
| 983 |
+
"step": 138
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.1440329218106995,
|
| 987 |
+
"grad_norm": 0.10248132676392517,
|
| 988 |
+
"learning_rate": 0.00015545682432067067,
|
| 989 |
+
"loss": 0.1235,
|
| 990 |
+
"step": 139
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.1522633744855968,
|
| 994 |
+
"grad_norm": 0.10737832299551002,
|
| 995 |
+
"learning_rate": 0.0001546523477511754,
|
| 996 |
+
"loss": 0.15,
|
| 997 |
+
"step": 140
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.1604938271604939,
|
| 1001 |
+
"grad_norm": 0.10807513117026502,
|
| 1002 |
+
"learning_rate": 0.00015384279578481221,
|
| 1003 |
+
"loss": 0.1302,
|
| 1004 |
+
"step": 141
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.168724279835391,
|
| 1008 |
+
"grad_norm": 0.11235063041052787,
|
| 1009 |
+
"learning_rate": 0.00015302824360220353,
|
| 1010 |
+
"loss": 0.1386,
|
| 1011 |
+
"step": 142
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.176954732510288,
|
| 1015 |
+
"grad_norm": 0.11303603244571206,
|
| 1016 |
+
"learning_rate": 0.00015220876684832638,
|
| 1017 |
+
"loss": 0.1354,
|
| 1018 |
+
"step": 143
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.1851851851851851,
|
| 1022 |
+
"grad_norm": 0.10834449382941344,
|
| 1023 |
+
"learning_rate": 0.0001513844416254879,
|
| 1024 |
+
"loss": 0.1329,
|
| 1025 |
+
"step": 144
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.1934156378600824,
|
| 1029 |
+
"grad_norm": 0.1079013280016857,
|
| 1030 |
+
"learning_rate": 0.00015055534448625766,
|
| 1031 |
+
"loss": 0.1395,
|
| 1032 |
+
"step": 145
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.2016460905349795,
|
| 1036 |
+
"grad_norm": 0.10261607273228789,
|
| 1037 |
+
"learning_rate": 0.00014972155242635852,
|
| 1038 |
+
"loss": 0.129,
|
| 1039 |
+
"step": 146
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.2098765432098766,
|
| 1043 |
+
"grad_norm": 0.11363708966450349,
|
| 1044 |
+
"learning_rate": 0.0001488831428775164,
|
| 1045 |
+
"loss": 0.1461,
|
| 1046 |
+
"step": 147
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.2181069958847737,
|
| 1050 |
+
"grad_norm": 0.11418227389762935,
|
| 1051 |
+
"learning_rate": 0.00014804019370026926,
|
| 1052 |
+
"loss": 0.1408,
|
| 1053 |
+
"step": 148
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.2263374485596708,
|
| 1057 |
+
"grad_norm": 0.11642847183223218,
|
| 1058 |
+
"learning_rate": 0.00014719278317673655,
|
| 1059 |
+
"loss": 0.1462,
|
| 1060 |
+
"step": 149
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.2345679012345678,
|
| 1064 |
+
"grad_norm": 0.11219874980224723,
|
| 1065 |
+
"learning_rate": 0.0001463409900033493,
|
| 1066 |
+
"loss": 0.1302,
|
| 1067 |
+
"step": 150
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.242798353909465,
|
| 1071 |
+
"grad_norm": 0.11039753961672125,
|
| 1072 |
+
"learning_rate": 0.00014548489328354195,
|
| 1073 |
+
"loss": 0.1349,
|
| 1074 |
+
"step": 151
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.2510288065843622,
|
| 1078 |
+
"grad_norm": 0.10772127499535779,
|
| 1079 |
+
"learning_rate": 0.00014462457252040607,
|
| 1080 |
+
"loss": 0.134,
|
| 1081 |
+
"step": 152
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.2592592592592593,
|
| 1085 |
+
"grad_norm": 0.10693684319005581,
|
| 1086 |
+
"learning_rate": 0.00014376010760930728,
|
| 1087 |
+
"loss": 0.1314,
|
| 1088 |
+
"step": 153
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.2674897119341564,
|
| 1092 |
+
"grad_norm": 0.10099951149470732,
|
| 1093 |
+
"learning_rate": 0.00014289157883046568,
|
| 1094 |
+
"loss": 0.1314,
|
| 1095 |
+
"step": 154
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.2757201646090535,
|
| 1099 |
+
"grad_norm": 0.10714073475937398,
|
| 1100 |
+
"learning_rate": 0.0001420190668415002,
|
| 1101 |
+
"loss": 0.1168,
|
| 1102 |
+
"step": 155
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.2839506172839505,
|
| 1106 |
+
"grad_norm": 0.11431029472007842,
|
| 1107 |
+
"learning_rate": 0.00014114265266993846,
|
| 1108 |
+
"loss": 0.1457,
|
| 1109 |
+
"step": 156
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.2921810699588478,
|
| 1113 |
+
"grad_norm": 0.11035801526331707,
|
| 1114 |
+
"learning_rate": 0.00014026241770569197,
|
| 1115 |
+
"loss": 0.1496,
|
| 1116 |
+
"step": 157
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.300411522633745,
|
| 1120 |
+
"grad_norm": 0.10419031063352954,
|
| 1121 |
+
"learning_rate": 0.00013937844369349734,
|
| 1122 |
+
"loss": 0.1323,
|
| 1123 |
+
"step": 158
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.308641975308642,
|
| 1127 |
+
"grad_norm": 0.10823681149718124,
|
| 1128 |
+
"learning_rate": 0.00013849081272532544,
|
| 1129 |
+
"loss": 0.1264,
|
| 1130 |
+
"step": 159
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.316872427983539,
|
| 1134 |
+
"grad_norm": 0.11224858342990347,
|
| 1135 |
+
"learning_rate": 0.00013759960723275732,
|
| 1136 |
+
"loss": 0.1494,
|
| 1137 |
+
"step": 160
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.3251028806584362,
|
| 1141 |
+
"grad_norm": 0.11222637320000997,
|
| 1142 |
+
"learning_rate": 0.00013670490997932922,
|
| 1143 |
+
"loss": 0.1446,
|
| 1144 |
+
"step": 161
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.3333333333333333,
|
| 1148 |
+
"grad_norm": 0.11350101469976674,
|
| 1149 |
+
"learning_rate": 0.00013580680405284664,
|
| 1150 |
+
"loss": 0.1501,
|
| 1151 |
+
"step": 162
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.3415637860082303,
|
| 1155 |
+
"grad_norm": 0.11216990215296897,
|
| 1156 |
+
"learning_rate": 0.00013490537285766808,
|
| 1157 |
+
"loss": 0.1518,
|
| 1158 |
+
"step": 163
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.3497942386831276,
|
| 1162 |
+
"grad_norm": 0.10385126129598882,
|
| 1163 |
+
"learning_rate": 0.00013400070010695966,
|
| 1164 |
+
"loss": 0.1326,
|
| 1165 |
+
"step": 164
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.3580246913580247,
|
| 1169 |
+
"grad_norm": 0.09801204121771982,
|
| 1170 |
+
"learning_rate": 0.00013309286981492085,
|
| 1171 |
+
"loss": 0.1385,
|
| 1172 |
+
"step": 165
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.3662551440329218,
|
| 1176 |
+
"grad_norm": 0.10515015522554948,
|
| 1177 |
+
"learning_rate": 0.00013218196628898233,
|
| 1178 |
+
"loss": 0.1435,
|
| 1179 |
+
"step": 166
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.374485596707819,
|
| 1183 |
+
"grad_norm": 0.11167870227165867,
|
| 1184 |
+
"learning_rate": 0.00013126807412197665,
|
| 1185 |
+
"loss": 0.1469,
|
| 1186 |
+
"step": 167
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.382716049382716,
|
| 1190 |
+
"grad_norm": 0.110692438020908,
|
| 1191 |
+
"learning_rate": 0.0001303512781842824,
|
| 1192 |
+
"loss": 0.1267,
|
| 1193 |
+
"step": 168
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.3909465020576133,
|
| 1197 |
+
"grad_norm": 0.11426160677474222,
|
| 1198 |
+
"learning_rate": 0.00012943166361594242,
|
| 1199 |
+
"loss": 0.1308,
|
| 1200 |
+
"step": 169
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.3991769547325104,
|
| 1204 |
+
"grad_norm": 0.11583690935003194,
|
| 1205 |
+
"learning_rate": 0.00012850931581875723,
|
| 1206 |
+
"loss": 0.1484,
|
| 1207 |
+
"step": 170
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.4074074074074074,
|
| 1211 |
+
"grad_norm": 0.10018605187361156,
|
| 1212 |
+
"learning_rate": 0.00012758432044835392,
|
| 1213 |
+
"loss": 0.141,
|
| 1214 |
+
"step": 171
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.4156378600823045,
|
| 1218 |
+
"grad_norm": 0.10604765462173282,
|
| 1219 |
+
"learning_rate": 0.0001266567634062317,
|
| 1220 |
+
"loss": 0.1291,
|
| 1221 |
+
"step": 172
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.4238683127572016,
|
| 1225 |
+
"grad_norm": 0.10158287607968718,
|
| 1226 |
+
"learning_rate": 0.0001257267308317845,
|
| 1227 |
+
"loss": 0.1276,
|
| 1228 |
+
"step": 173
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.4320987654320987,
|
| 1232 |
+
"grad_norm": 0.10299808028319524,
|
| 1233 |
+
"learning_rate": 0.00012479430909430108,
|
| 1234 |
+
"loss": 0.1317,
|
| 1235 |
+
"step": 174
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.4403292181069958,
|
| 1239 |
+
"grad_norm": 0.1118568576111719,
|
| 1240 |
+
"learning_rate": 0.00012385958478494487,
|
| 1241 |
+
"loss": 0.1288,
|
| 1242 |
+
"step": 175
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.448559670781893,
|
| 1246 |
+
"grad_norm": 0.10353944126895748,
|
| 1247 |
+
"learning_rate": 0.00012292264470871182,
|
| 1248 |
+
"loss": 0.1175,
|
| 1249 |
+
"step": 176
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.4567901234567902,
|
| 1253 |
+
"grad_norm": 0.11318325883515341,
|
| 1254 |
+
"learning_rate": 0.00012198357587636957,
|
| 1255 |
+
"loss": 0.1304,
|
| 1256 |
+
"step": 177
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.4650205761316872,
|
| 1260 |
+
"grad_norm": 0.11698708468284778,
|
| 1261 |
+
"learning_rate": 0.00012104246549637683,
|
| 1262 |
+
"loss": 0.143,
|
| 1263 |
+
"step": 178
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.4732510288065843,
|
| 1267 |
+
"grad_norm": 0.12408074070049697,
|
| 1268 |
+
"learning_rate": 0.00012009940096678452,
|
| 1269 |
+
"loss": 0.1583,
|
| 1270 |
+
"step": 179
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.4814814814814814,
|
| 1274 |
+
"grad_norm": 0.115849428171122,
|
| 1275 |
+
"learning_rate": 0.00011915446986711953,
|
| 1276 |
+
"loss": 0.1401,
|
| 1277 |
+
"step": 180
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.4897119341563787,
|
| 1281 |
+
"grad_norm": 0.10498841318355592,
|
| 1282 |
+
"learning_rate": 0.00011820775995025147,
|
| 1283 |
+
"loss": 0.1404,
|
| 1284 |
+
"step": 181
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.4979423868312758,
|
| 1288 |
+
"grad_norm": 0.11284118297477287,
|
| 1289 |
+
"learning_rate": 0.0001172593591342432,
|
| 1290 |
+
"loss": 0.1479,
|
| 1291 |
+
"step": 182
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.5061728395061729,
|
| 1295 |
+
"grad_norm": 0.1063063021523787,
|
| 1296 |
+
"learning_rate": 0.00011630935549418627,
|
| 1297 |
+
"loss": 0.1223,
|
| 1298 |
+
"step": 183
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.51440329218107,
|
| 1302 |
+
"grad_norm": 0.09875055061391125,
|
| 1303 |
+
"learning_rate": 0.00011535783725402163,
|
| 1304 |
+
"loss": 0.1177,
|
| 1305 |
+
"step": 184
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.522633744855967,
|
| 1309 |
+
"grad_norm": 0.10114695106701617,
|
| 1310 |
+
"learning_rate": 0.00011440489277834645,
|
| 1311 |
+
"loss": 0.1403,
|
| 1312 |
+
"step": 185
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.5308641975308643,
|
| 1316 |
+
"grad_norm": 0.09017689726172246,
|
| 1317 |
+
"learning_rate": 0.0001134506105642081,
|
| 1318 |
+
"loss": 0.1114,
|
| 1319 |
+
"step": 186
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.5390946502057612,
|
| 1323 |
+
"grad_norm": 0.09727031699745013,
|
| 1324 |
+
"learning_rate": 0.00011249507923288562,
|
| 1325 |
+
"loss": 0.1368,
|
| 1326 |
+
"step": 187
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.5473251028806585,
|
| 1330 |
+
"grad_norm": 0.10131253354694934,
|
| 1331 |
+
"learning_rate": 0.0001115383875216598,
|
| 1332 |
+
"loss": 0.1278,
|
| 1333 |
+
"step": 188
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.5555555555555556,
|
| 1337 |
+
"grad_norm": 0.09766756433474798,
|
| 1338 |
+
"learning_rate": 0.00011058062427557229,
|
| 1339 |
+
"loss": 0.1284,
|
| 1340 |
+
"step": 189
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.5637860082304527,
|
| 1344 |
+
"grad_norm": 0.09911328195816287,
|
| 1345 |
+
"learning_rate": 0.00010962187843917497,
|
| 1346 |
+
"loss": 0.1284,
|
| 1347 |
+
"step": 190
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.5720164609053497,
|
| 1351 |
+
"grad_norm": 0.10809617890378591,
|
| 1352 |
+
"learning_rate": 0.0001086622390482699,
|
| 1353 |
+
"loss": 0.1423,
|
| 1354 |
+
"step": 191
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.5802469135802468,
|
| 1358 |
+
"grad_norm": 0.09456158065899406,
|
| 1359 |
+
"learning_rate": 0.00010770179522164079,
|
| 1360 |
+
"loss": 0.1317,
|
| 1361 |
+
"step": 192
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.5884773662551441,
|
| 1365 |
+
"grad_norm": 0.09799719160217063,
|
| 1366 |
+
"learning_rate": 0.0001067406361527768,
|
| 1367 |
+
"loss": 0.1356,
|
| 1368 |
+
"step": 193
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.596707818930041,
|
| 1372 |
+
"grad_norm": 0.09992479118544549,
|
| 1373 |
+
"learning_rate": 0.00010577885110158958,
|
| 1374 |
+
"loss": 0.1292,
|
| 1375 |
+
"step": 194
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.6049382716049383,
|
| 1379 |
+
"grad_norm": 0.09693089475338096,
|
| 1380 |
+
"learning_rate": 0.00010481652938612374,
|
| 1381 |
+
"loss": 0.1391,
|
| 1382 |
+
"step": 195
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.6131687242798354,
|
| 1386 |
+
"grad_norm": 0.09217578544631694,
|
| 1387 |
+
"learning_rate": 0.00010385376037426226,
|
| 1388 |
+
"loss": 0.1152,
|
| 1389 |
+
"step": 196
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.6213991769547325,
|
| 1393 |
+
"grad_norm": 0.09249375316990154,
|
| 1394 |
+
"learning_rate": 0.00010289063347542726,
|
| 1395 |
+
"loss": 0.1154,
|
| 1396 |
+
"step": 197
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.6296296296296298,
|
| 1400 |
+
"grad_norm": 0.09641811645171969,
|
| 1401 |
+
"learning_rate": 0.00010192723813227672,
|
| 1402 |
+
"loss": 0.1182,
|
| 1403 |
+
"step": 198
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.6378600823045266,
|
| 1407 |
+
"grad_norm": 0.09748485166957398,
|
| 1408 |
+
"learning_rate": 0.00010096366381239808,
|
| 1409 |
+
"loss": 0.1338,
|
| 1410 |
+
"step": 199
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.646090534979424,
|
| 1414 |
+
"grad_norm": 0.09383283234337292,
|
| 1415 |
+
"learning_rate": 0.0001,
|
| 1416 |
+
"loss": 0.1235,
|
| 1417 |
+
"step": 200
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.654320987654321,
|
| 1421 |
+
"grad_norm": 0.09670711594290475,
|
| 1422 |
+
"learning_rate": 9.903633618760195e-05,
|
| 1423 |
+
"loss": 0.1222,
|
| 1424 |
+
"step": 201
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.662551440329218,
|
| 1428 |
+
"grad_norm": 0.09682675756651625,
|
| 1429 |
+
"learning_rate": 9.807276186772333e-05,
|
| 1430 |
+
"loss": 0.1271,
|
| 1431 |
+
"step": 202
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.6707818930041154,
|
| 1435 |
+
"grad_norm": 0.09455470748347092,
|
| 1436 |
+
"learning_rate": 9.710936652457276e-05,
|
| 1437 |
+
"loss": 0.1217,
|
| 1438 |
+
"step": 203
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.6790123456790123,
|
| 1442 |
+
"grad_norm": 0.10988834676545467,
|
| 1443 |
+
"learning_rate": 9.614623962573776e-05,
|
| 1444 |
+
"loss": 0.1288,
|
| 1445 |
+
"step": 204
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.6872427983539096,
|
| 1449 |
+
"grad_norm": 0.10415165974330459,
|
| 1450 |
+
"learning_rate": 9.518347061387628e-05,
|
| 1451 |
+
"loss": 0.1307,
|
| 1452 |
+
"step": 205
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.6954732510288066,
|
| 1456 |
+
"grad_norm": 0.10985786659106267,
|
| 1457 |
+
"learning_rate": 9.422114889841044e-05,
|
| 1458 |
+
"loss": 0.1405,
|
| 1459 |
+
"step": 206
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.7037037037037037,
|
| 1463 |
+
"grad_norm": 0.09921801439791741,
|
| 1464 |
+
"learning_rate": 9.325936384722321e-05,
|
| 1465 |
+
"loss": 0.1421,
|
| 1466 |
+
"step": 207
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.7119341563786008,
|
| 1470 |
+
"grad_norm": 0.09695464903081527,
|
| 1471 |
+
"learning_rate": 9.229820477835927e-05,
|
| 1472 |
+
"loss": 0.1261,
|
| 1473 |
+
"step": 208
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.7201646090534979,
|
| 1477 |
+
"grad_norm": 0.09747696306459716,
|
| 1478 |
+
"learning_rate": 9.133776095173015e-05,
|
| 1479 |
+
"loss": 0.1295,
|
| 1480 |
+
"step": 209
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.7283950617283952,
|
| 1484 |
+
"grad_norm": 0.09028526494291234,
|
| 1485 |
+
"learning_rate": 9.037812156082504e-05,
|
| 1486 |
+
"loss": 0.1148,
|
| 1487 |
+
"step": 210
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.736625514403292,
|
| 1491 |
+
"grad_norm": 0.11512926087416764,
|
| 1492 |
+
"learning_rate": 8.941937572442773e-05,
|
| 1493 |
+
"loss": 0.1385,
|
| 1494 |
+
"step": 211
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.7448559670781894,
|
| 1498 |
+
"grad_norm": 0.09719378966338557,
|
| 1499 |
+
"learning_rate": 8.846161247834024e-05,
|
| 1500 |
+
"loss": 0.1308,
|
| 1501 |
+
"step": 212
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.7530864197530864,
|
| 1505 |
+
"grad_norm": 0.09008557356074057,
|
| 1506 |
+
"learning_rate": 8.750492076711439e-05,
|
| 1507 |
+
"loss": 0.1142,
|
| 1508 |
+
"step": 213
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.7613168724279835,
|
| 1512 |
+
"grad_norm": 0.10577290101156246,
|
| 1513 |
+
"learning_rate": 8.654938943579194e-05,
|
| 1514 |
+
"loss": 0.1409,
|
| 1515 |
+
"step": 214
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.7695473251028808,
|
| 1519 |
+
"grad_norm": 0.10122461461639144,
|
| 1520 |
+
"learning_rate": 8.55951072216536e-05,
|
| 1521 |
+
"loss": 0.1314,
|
| 1522 |
+
"step": 215
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.7777777777777777,
|
| 1526 |
+
"grad_norm": 0.09823573980962558,
|
| 1527 |
+
"learning_rate": 8.464216274597838e-05,
|
| 1528 |
+
"loss": 0.1219,
|
| 1529 |
+
"step": 216
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.786008230452675,
|
| 1533 |
+
"grad_norm": 0.10401921170876789,
|
| 1534 |
+
"learning_rate": 8.369064450581373e-05,
|
| 1535 |
+
"loss": 0.144,
|
| 1536 |
+
"step": 217
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.794238683127572,
|
| 1540 |
+
"grad_norm": 0.09078999043742002,
|
| 1541 |
+
"learning_rate": 8.274064086575681e-05,
|
| 1542 |
+
"loss": 0.1146,
|
| 1543 |
+
"step": 218
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.8024691358024691,
|
| 1547 |
+
"grad_norm": 0.09290736424935679,
|
| 1548 |
+
"learning_rate": 8.179224004974857e-05,
|
| 1549 |
+
"loss": 0.1338,
|
| 1550 |
+
"step": 219
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.8106995884773662,
|
| 1554 |
+
"grad_norm": 0.10089995497723929,
|
| 1555 |
+
"learning_rate": 8.084553013288048e-05,
|
| 1556 |
+
"loss": 0.1299,
|
| 1557 |
+
"step": 220
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.8189300411522633,
|
| 1561 |
+
"grad_norm": 0.11086354657368475,
|
| 1562 |
+
"learning_rate": 7.990059903321553e-05,
|
| 1563 |
+
"loss": 0.1413,
|
| 1564 |
+
"step": 221
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.8271604938271606,
|
| 1568 |
+
"grad_norm": 0.10239856014254489,
|
| 1569 |
+
"learning_rate": 7.89575345036232e-05,
|
| 1570 |
+
"loss": 0.1357,
|
| 1571 |
+
"step": 222
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.8353909465020575,
|
| 1575 |
+
"grad_norm": 0.10469014425315878,
|
| 1576 |
+
"learning_rate": 7.801642412363041e-05,
|
| 1577 |
+
"loss": 0.1386,
|
| 1578 |
+
"step": 223
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.8436213991769548,
|
| 1582 |
+
"grad_norm": 0.09567702676769906,
|
| 1583 |
+
"learning_rate": 7.707735529128819e-05,
|
| 1584 |
+
"loss": 0.1191,
|
| 1585 |
+
"step": 224
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.8518518518518519,
|
| 1589 |
+
"grad_norm": 0.09887833758041985,
|
| 1590 |
+
"learning_rate": 7.614041521505517e-05,
|
| 1591 |
+
"loss": 0.1219,
|
| 1592 |
+
"step": 225
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.860082304526749,
|
| 1596 |
+
"grad_norm": 0.11752510500534394,
|
| 1597 |
+
"learning_rate": 7.520569090569893e-05,
|
| 1598 |
+
"loss": 0.1436,
|
| 1599 |
+
"step": 226
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.8683127572016462,
|
| 1603 |
+
"grad_norm": 0.10388061868632578,
|
| 1604 |
+
"learning_rate": 7.427326916821557e-05,
|
| 1605 |
+
"loss": 0.1521,
|
| 1606 |
+
"step": 227
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.876543209876543,
|
| 1610 |
+
"grad_norm": 0.09472507284073291,
|
| 1611 |
+
"learning_rate": 7.334323659376829e-05,
|
| 1612 |
+
"loss": 0.1309,
|
| 1613 |
+
"step": 228
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.8847736625514404,
|
| 1617 |
+
"grad_norm": 0.09368373509928322,
|
| 1618 |
+
"learning_rate": 7.24156795516461e-05,
|
| 1619 |
+
"loss": 0.1226,
|
| 1620 |
+
"step": 229
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.8930041152263375,
|
| 1624 |
+
"grad_norm": 0.09798127382996634,
|
| 1625 |
+
"learning_rate": 7.149068418124281e-05,
|
| 1626 |
+
"loss": 0.134,
|
| 1627 |
+
"step": 230
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.9012345679012346,
|
| 1631 |
+
"grad_norm": 0.09780244211128043,
|
| 1632 |
+
"learning_rate": 7.056833638405762e-05,
|
| 1633 |
+
"loss": 0.1139,
|
| 1634 |
+
"step": 231
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.9094650205761317,
|
| 1638 |
+
"grad_norm": 0.09955863946262253,
|
| 1639 |
+
"learning_rate": 6.964872181571764e-05,
|
| 1640 |
+
"loss": 0.1372,
|
| 1641 |
+
"step": 232
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.9176954732510287,
|
| 1645 |
+
"grad_norm": 0.0999251460303617,
|
| 1646 |
+
"learning_rate": 6.87319258780234e-05,
|
| 1647 |
+
"loss": 0.1302,
|
| 1648 |
+
"step": 233
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.925925925925926,
|
| 1652 |
+
"grad_norm": 0.10537500302863075,
|
| 1653 |
+
"learning_rate": 6.781803371101774e-05,
|
| 1654 |
+
"loss": 0.1361,
|
| 1655 |
+
"step": 234
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.934156378600823,
|
| 1659 |
+
"grad_norm": 0.10646426579724969,
|
| 1660 |
+
"learning_rate": 6.690713018507918e-05,
|
| 1661 |
+
"loss": 0.1382,
|
| 1662 |
+
"step": 235
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.9423868312757202,
|
| 1666 |
+
"grad_norm": 0.09815506133539018,
|
| 1667 |
+
"learning_rate": 6.599929989304035e-05,
|
| 1668 |
+
"loss": 0.1248,
|
| 1669 |
+
"step": 236
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.9506172839506173,
|
| 1673 |
+
"grad_norm": 0.10476726048050965,
|
| 1674 |
+
"learning_rate": 6.509462714233195e-05,
|
| 1675 |
+
"loss": 0.139,
|
| 1676 |
+
"step": 237
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.9588477366255144,
|
| 1680 |
+
"grad_norm": 0.10022243654627384,
|
| 1681 |
+
"learning_rate": 6.419319594715339e-05,
|
| 1682 |
+
"loss": 0.1285,
|
| 1683 |
+
"step": 238
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.9670781893004117,
|
| 1687 |
+
"grad_norm": 0.09626907084774464,
|
| 1688 |
+
"learning_rate": 6.32950900206708e-05,
|
| 1689 |
+
"loss": 0.1345,
|
| 1690 |
+
"step": 239
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.9753086419753085,
|
| 1694 |
+
"grad_norm": 0.09684684548905444,
|
| 1695 |
+
"learning_rate": 6.240039276724272e-05,
|
| 1696 |
+
"loss": 0.1338,
|
| 1697 |
+
"step": 240
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.9835390946502058,
|
| 1701 |
+
"grad_norm": 0.09617321255871392,
|
| 1702 |
+
"learning_rate": 6.150918727467455e-05,
|
| 1703 |
+
"loss": 0.1475,
|
| 1704 |
+
"step": 241
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.991769547325103,
|
| 1708 |
+
"grad_norm": 0.10288685971512879,
|
| 1709 |
+
"learning_rate": 6.062155630650265e-05,
|
| 1710 |
+
"loss": 0.13,
|
| 1711 |
+
"step": 242
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 2.0,
|
| 1715 |
+
"grad_norm": 0.09997416312505185,
|
| 1716 |
+
"learning_rate": 5.973758229430806e-05,
|
| 1717 |
+
"loss": 0.1282,
|
| 1718 |
+
"step": 243
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 2.0,
|
| 1722 |
+
"eval_loss": 0.14014942944049835,
|
| 1723 |
+
"eval_runtime": 21.1655,
|
| 1724 |
+
"eval_samples_per_second": 38.601,
|
| 1725 |
+
"eval_steps_per_second": 1.228,
|
| 1726 |
+
"step": 243
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 2.0082304526748973,
|
| 1730 |
+
"grad_norm": 0.0846538233930147,
|
| 1731 |
+
"learning_rate": 5.885734733006154e-05,
|
| 1732 |
+
"loss": 0.0975,
|
| 1733 |
+
"step": 244
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 2.016460905349794,
|
| 1737 |
+
"grad_norm": 0.09549609421876461,
|
| 1738 |
+
"learning_rate": 5.798093315849984e-05,
|
| 1739 |
+
"loss": 0.1102,
|
| 1740 |
+
"step": 245
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 2.0246913580246915,
|
| 1744 |
+
"grad_norm": 0.09559068048673255,
|
| 1745 |
+
"learning_rate": 5.710842116953438e-05,
|
| 1746 |
+
"loss": 0.1025,
|
| 1747 |
+
"step": 246
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 2.0329218106995883,
|
| 1751 |
+
"grad_norm": 0.10047776150134546,
|
| 1752 |
+
"learning_rate": 5.623989239069275e-05,
|
| 1753 |
+
"loss": 0.1167,
|
| 1754 |
+
"step": 247
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 2.0411522633744856,
|
| 1758 |
+
"grad_norm": 0.10670187619168023,
|
| 1759 |
+
"learning_rate": 5.537542747959394e-05,
|
| 1760 |
+
"loss": 0.1115,
|
| 1761 |
+
"step": 248
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 2.049382716049383,
|
| 1765 |
+
"grad_norm": 0.10560096207358394,
|
| 1766 |
+
"learning_rate": 5.451510671645807e-05,
|
| 1767 |
+
"loss": 0.117,
|
| 1768 |
+
"step": 249
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 2.05761316872428,
|
| 1772 |
+
"grad_norm": 0.10122771696509619,
|
| 1773 |
+
"learning_rate": 5.36590099966507e-05,
|
| 1774 |
+
"loss": 0.1253,
|
| 1775 |
+
"step": 250
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 2.065843621399177,
|
| 1779 |
+
"grad_norm": 0.09738370841133107,
|
| 1780 |
+
"learning_rate": 5.2807216823263484e-05,
|
| 1781 |
+
"loss": 0.1188,
|
| 1782 |
+
"step": 251
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 2.074074074074074,
|
| 1786 |
+
"grad_norm": 0.09886958144813014,
|
| 1787 |
+
"learning_rate": 5.1959806299730774e-05,
|
| 1788 |
+
"loss": 0.1237,
|
| 1789 |
+
"step": 252
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 2.0823045267489713,
|
| 1793 |
+
"grad_norm": 0.10219049090054223,
|
| 1794 |
+
"learning_rate": 5.111685712248364e-05,
|
| 1795 |
+
"loss": 0.1137,
|
| 1796 |
+
"step": 253
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 2.090534979423868,
|
| 1800 |
+
"grad_norm": 0.09836202650897583,
|
| 1801 |
+
"learning_rate": 5.0278447573641495e-05,
|
| 1802 |
+
"loss": 0.1196,
|
| 1803 |
+
"step": 254
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 2.0987654320987654,
|
| 1807 |
+
"grad_norm": 0.09213881530201376,
|
| 1808 |
+
"learning_rate": 4.944465551374238e-05,
|
| 1809 |
+
"loss": 0.0999,
|
| 1810 |
+
"step": 255
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 2.1069958847736627,
|
| 1814 |
+
"grad_norm": 0.09541725924400123,
|
| 1815 |
+
"learning_rate": 4.861555837451213e-05,
|
| 1816 |
+
"loss": 0.1042,
|
| 1817 |
+
"step": 256
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 2.1152263374485596,
|
| 1821 |
+
"grad_norm": 0.09856792963956205,
|
| 1822 |
+
"learning_rate": 4.779123315167362e-05,
|
| 1823 |
+
"loss": 0.1139,
|
| 1824 |
+
"step": 257
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 2.123456790123457,
|
| 1828 |
+
"grad_norm": 0.09596664358323789,
|
| 1829 |
+
"learning_rate": 4.6971756397796504e-05,
|
| 1830 |
+
"loss": 0.1127,
|
| 1831 |
+
"step": 258
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 2.1316872427983538,
|
| 1835 |
+
"grad_norm": 0.10398664110407933,
|
| 1836 |
+
"learning_rate": 4.61572042151878e-05,
|
| 1837 |
+
"loss": 0.1226,
|
| 1838 |
+
"step": 259
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 2.139917695473251,
|
| 1842 |
+
"grad_norm": 0.09339956051495163,
|
| 1843 |
+
"learning_rate": 4.5347652248824624e-05,
|
| 1844 |
+
"loss": 0.1099,
|
| 1845 |
+
"step": 260
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 2.148148148148148,
|
| 1849 |
+
"grad_norm": 0.09890330904002925,
|
| 1850 |
+
"learning_rate": 4.4543175679329344e-05,
|
| 1851 |
+
"loss": 0.1195,
|
| 1852 |
+
"step": 261
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 2.156378600823045,
|
| 1856 |
+
"grad_norm": 0.09997598174399476,
|
| 1857 |
+
"learning_rate": 4.3743849215987595e-05,
|
| 1858 |
+
"loss": 0.1128,
|
| 1859 |
+
"step": 262
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 2.1646090534979425,
|
| 1863 |
+
"grad_norm": 0.09613540261918692,
|
| 1864 |
+
"learning_rate": 4.294974708981041e-05,
|
| 1865 |
+
"loss": 0.11,
|
| 1866 |
+
"step": 263
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 2.1728395061728394,
|
| 1870 |
+
"grad_norm": 0.09811570643997117,
|
| 1871 |
+
"learning_rate": 4.216094304664056e-05,
|
| 1872 |
+
"loss": 0.1221,
|
| 1873 |
+
"step": 264
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 2.1810699588477367,
|
| 1877 |
+
"grad_norm": 0.10130508026732313,
|
| 1878 |
+
"learning_rate": 4.137751034030399e-05,
|
| 1879 |
+
"loss": 0.1147,
|
| 1880 |
+
"step": 265
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 2.1893004115226335,
|
| 1884 |
+
"grad_norm": 0.10155247373344696,
|
| 1885 |
+
"learning_rate": 4.059952172580694e-05,
|
| 1886 |
+
"loss": 0.1258,
|
| 1887 |
+
"step": 266
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 2.197530864197531,
|
| 1891 |
+
"grad_norm": 0.10114305608112335,
|
| 1892 |
+
"learning_rate": 3.982704945257957e-05,
|
| 1893 |
+
"loss": 0.1125,
|
| 1894 |
+
"step": 267
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 2.205761316872428,
|
| 1898 |
+
"grad_norm": 0.09834925536076868,
|
| 1899 |
+
"learning_rate": 3.906016525776611e-05,
|
| 1900 |
+
"loss": 0.1178,
|
| 1901 |
+
"step": 268
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 2.213991769547325,
|
| 1905 |
+
"grad_norm": 0.10693349155074929,
|
| 1906 |
+
"learning_rate": 3.829894035956306e-05,
|
| 1907 |
+
"loss": 0.125,
|
| 1908 |
+
"step": 269
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 2.2222222222222223,
|
| 1912 |
+
"grad_norm": 0.09467414992649671,
|
| 1913 |
+
"learning_rate": 3.7543445450605285e-05,
|
| 1914 |
+
"loss": 0.1054,
|
| 1915 |
+
"step": 270
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 2.230452674897119,
|
| 1919 |
+
"grad_norm": 0.09826633722499567,
|
| 1920 |
+
"learning_rate": 3.6793750691400994e-05,
|
| 1921 |
+
"loss": 0.1066,
|
| 1922 |
+
"step": 271
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 2.2386831275720165,
|
| 1926 |
+
"grad_norm": 0.10348161204951604,
|
| 1927 |
+
"learning_rate": 3.6049925703816214e-05,
|
| 1928 |
+
"loss": 0.1144,
|
| 1929 |
+
"step": 272
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 2.246913580246914,
|
| 1933 |
+
"grad_norm": 0.10479446585724977,
|
| 1934 |
+
"learning_rate": 3.53120395646092e-05,
|
| 1935 |
+
"loss": 0.1135,
|
| 1936 |
+
"step": 273
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 2.2551440329218106,
|
| 1940 |
+
"grad_norm": 0.0951654749026979,
|
| 1941 |
+
"learning_rate": 3.458016079901544e-05,
|
| 1942 |
+
"loss": 0.1074,
|
| 1943 |
+
"step": 274
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 2.263374485596708,
|
| 1947 |
+
"grad_norm": 0.11076793087444725,
|
| 1948 |
+
"learning_rate": 3.38543573743839e-05,
|
| 1949 |
+
"loss": 0.1299,
|
| 1950 |
+
"step": 275
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 2.271604938271605,
|
| 1954 |
+
"grad_norm": 0.0946857687815803,
|
| 1955 |
+
"learning_rate": 3.3134696693865316e-05,
|
| 1956 |
+
"loss": 0.106,
|
| 1957 |
+
"step": 276
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 2.279835390946502,
|
| 1961 |
+
"grad_norm": 0.09607169531950183,
|
| 1962 |
+
"learning_rate": 3.242124559015234e-05,
|
| 1963 |
+
"loss": 0.0966,
|
| 1964 |
+
"step": 277
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 2.288065843621399,
|
| 1968 |
+
"grad_norm": 0.10167592887197249,
|
| 1969 |
+
"learning_rate": 3.171407031927325e-05,
|
| 1970 |
+
"loss": 0.1162,
|
| 1971 |
+
"step": 278
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 2.2962962962962963,
|
| 1975 |
+
"grad_norm": 0.09726294827804387,
|
| 1976 |
+
"learning_rate": 3.101323655443882e-05,
|
| 1977 |
+
"loss": 0.1139,
|
| 1978 |
+
"step": 279
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 2.3045267489711936,
|
| 1982 |
+
"grad_norm": 0.09582434907073016,
|
| 1983 |
+
"learning_rate": 3.031880937994359e-05,
|
| 1984 |
+
"loss": 0.1084,
|
| 1985 |
+
"step": 280
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 2.3127572016460904,
|
| 1989 |
+
"grad_norm": 0.11021289451624465,
|
| 1990 |
+
"learning_rate": 2.9630853285121508e-05,
|
| 1991 |
+
"loss": 0.1231,
|
| 1992 |
+
"step": 281
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 2.3209876543209877,
|
| 1996 |
+
"grad_norm": 0.09391459178579925,
|
| 1997 |
+
"learning_rate": 2.894943215835708e-05,
|
| 1998 |
+
"loss": 0.1047,
|
| 1999 |
+
"step": 282
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 2.3292181069958846,
|
| 2003 |
+
"grad_norm": 0.09899700696295954,
|
| 2004 |
+
"learning_rate": 2.827460928115232e-05,
|
| 2005 |
+
"loss": 0.1101,
|
| 2006 |
+
"step": 283
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 2.337448559670782,
|
| 2010 |
+
"grad_norm": 0.09901505034589045,
|
| 2011 |
+
"learning_rate": 2.7606447322249872e-05,
|
| 2012 |
+
"loss": 0.1067,
|
| 2013 |
+
"step": 284
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 2.3456790123456788,
|
| 2017 |
+
"grad_norm": 0.10420896589962987,
|
| 2018 |
+
"learning_rate": 2.6945008331813226e-05,
|
| 2019 |
+
"loss": 0.1282,
|
| 2020 |
+
"step": 285
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 2.353909465020576,
|
| 2024 |
+
"grad_norm": 0.09729393595160353,
|
| 2025 |
+
"learning_rate": 2.629035373566433e-05,
|
| 2026 |
+
"loss": 0.1023,
|
| 2027 |
+
"step": 286
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 2.3621399176954734,
|
| 2031 |
+
"grad_norm": 0.09829406149572398,
|
| 2032 |
+
"learning_rate": 2.5642544329579088e-05,
|
| 2033 |
+
"loss": 0.1155,
|
| 2034 |
+
"step": 287
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 2.3703703703703702,
|
| 2038 |
+
"grad_norm": 0.10275430043325989,
|
| 2039 |
+
"learning_rate": 2.500164027364147e-05,
|
| 2040 |
+
"loss": 0.1252,
|
| 2041 |
+
"step": 288
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 2.3786008230452675,
|
| 2045 |
+
"grad_norm": 0.10367750094719844,
|
| 2046 |
+
"learning_rate": 2.4367701086656624e-05,
|
| 2047 |
+
"loss": 0.1166,
|
| 2048 |
+
"step": 289
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 2.386831275720165,
|
| 2052 |
+
"grad_norm": 0.10274988439011652,
|
| 2053 |
+
"learning_rate": 2.3740785640623643e-05,
|
| 2054 |
+
"loss": 0.1169,
|
| 2055 |
+
"step": 290
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 2.3950617283950617,
|
| 2059 |
+
"grad_norm": 0.09057411016580416,
|
| 2060 |
+
"learning_rate": 2.312095215526814e-05,
|
| 2061 |
+
"loss": 0.1026,
|
| 2062 |
+
"step": 291
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 2.403292181069959,
|
| 2066 |
+
"grad_norm": 0.104243770759663,
|
| 2067 |
+
"learning_rate": 2.2508258192635612e-05,
|
| 2068 |
+
"loss": 0.1127,
|
| 2069 |
+
"step": 292
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 2.411522633744856,
|
| 2073 |
+
"grad_norm": 0.09595429188209968,
|
| 2074 |
+
"learning_rate": 2.1902760651745958e-05,
|
| 2075 |
+
"loss": 0.1117,
|
| 2076 |
+
"step": 293
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 2.419753086419753,
|
| 2080 |
+
"grad_norm": 0.10143248145662324,
|
| 2081 |
+
"learning_rate": 2.1304515763309253e-05,
|
| 2082 |
+
"loss": 0.111,
|
| 2083 |
+
"step": 294
|
| 2084 |
+
},
|
| 2085 |
+
{
|
| 2086 |
+
"epoch": 2.42798353909465,
|
| 2087 |
+
"grad_norm": 0.1074361000716321,
|
| 2088 |
+
"learning_rate": 2.0713579084503876e-05,
|
| 2089 |
+
"loss": 0.1222,
|
| 2090 |
+
"step": 295
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 2.4362139917695473,
|
| 2094 |
+
"grad_norm": 0.09389635392057138,
|
| 2095 |
+
"learning_rate": 2.013000549381706e-05,
|
| 2096 |
+
"loss": 0.1009,
|
| 2097 |
+
"step": 296
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 2.4444444444444446,
|
| 2101 |
+
"grad_norm": 0.0950080424907371,
|
| 2102 |
+
"learning_rate": 1.9553849185948512e-05,
|
| 2103 |
+
"loss": 0.1111,
|
| 2104 |
+
"step": 297
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 2.4526748971193415,
|
| 2108 |
+
"grad_norm": 0.09761821744754258,
|
| 2109 |
+
"learning_rate": 1.8985163666777473e-05,
|
| 2110 |
+
"loss": 0.1013,
|
| 2111 |
+
"step": 298
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 2.460905349794239,
|
| 2115 |
+
"grad_norm": 0.10939996129330187,
|
| 2116 |
+
"learning_rate": 1.8424001748393905e-05,
|
| 2117 |
+
"loss": 0.1215,
|
| 2118 |
+
"step": 299
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"epoch": 2.4691358024691357,
|
| 2122 |
+
"grad_norm": 0.09848524806568702,
|
| 2123 |
+
"learning_rate": 1.787041554419381e-05,
|
| 2124 |
+
"loss": 0.1,
|
| 2125 |
+
"step": 300
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 2.477366255144033,
|
| 2129 |
+
"grad_norm": 0.10208382321734329,
|
| 2130 |
+
"learning_rate": 1.7324456464039752e-05,
|
| 2131 |
+
"loss": 0.1156,
|
| 2132 |
+
"step": 301
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 2.48559670781893,
|
| 2136 |
+
"grad_norm": 0.10860830565047937,
|
| 2137 |
+
"learning_rate": 1.6786175209486566e-05,
|
| 2138 |
+
"loss": 0.1135,
|
| 2139 |
+
"step": 302
|
| 2140 |
+
},
|
| 2141 |
+
{
|
| 2142 |
+
"epoch": 2.493827160493827,
|
| 2143 |
+
"grad_norm": 0.11254066618708665,
|
| 2144 |
+
"learning_rate": 1.6255621769072805e-05,
|
| 2145 |
+
"loss": 0.1202,
|
| 2146 |
+
"step": 303
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 2.5020576131687244,
|
| 2150 |
+
"grad_norm": 0.09949710536532337,
|
| 2151 |
+
"learning_rate": 1.5732845413678477e-05,
|
| 2152 |
+
"loss": 0.108,
|
| 2153 |
+
"step": 304
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 2.5102880658436213,
|
| 2157 |
+
"grad_norm": 0.09871004589964077,
|
| 2158 |
+
"learning_rate": 1.521789469194952e-05,
|
| 2159 |
+
"loss": 0.1048,
|
| 2160 |
+
"step": 305
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 2.5185185185185186,
|
| 2164 |
+
"grad_norm": 0.09509589762300931,
|
| 2165 |
+
"learning_rate": 1.4710817425789014e-05,
|
| 2166 |
+
"loss": 0.1108,
|
| 2167 |
+
"step": 306
|
| 2168 |
+
},
|
| 2169 |
+
{
|
| 2170 |
+
"epoch": 2.526748971193416,
|
| 2171 |
+
"grad_norm": 0.09874336616077671,
|
| 2172 |
+
"learning_rate": 1.4211660705916285e-05,
|
| 2173 |
+
"loss": 0.1075,
|
| 2174 |
+
"step": 307
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"epoch": 2.5349794238683128,
|
| 2178 |
+
"grad_norm": 0.10610686628994136,
|
| 2179 |
+
"learning_rate": 1.3720470887493719e-05,
|
| 2180 |
+
"loss": 0.114,
|
| 2181 |
+
"step": 308
|
| 2182 |
+
},
|
| 2183 |
+
{
|
| 2184 |
+
"epoch": 2.5432098765432096,
|
| 2185 |
+
"grad_norm": 0.09371419408053047,
|
| 2186 |
+
"learning_rate": 1.3237293585821786e-05,
|
| 2187 |
+
"loss": 0.1,
|
| 2188 |
+
"step": 309
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"epoch": 2.551440329218107,
|
| 2192 |
+
"grad_norm": 0.097056043473284,
|
| 2193 |
+
"learning_rate": 1.2762173672102996e-05,
|
| 2194 |
+
"loss": 0.1091,
|
| 2195 |
+
"step": 310
|
| 2196 |
+
},
|
| 2197 |
+
{
|
| 2198 |
+
"epoch": 2.5596707818930042,
|
| 2199 |
+
"grad_norm": 0.10289500706688874,
|
| 2200 |
+
"learning_rate": 1.2295155269274827e-05,
|
| 2201 |
+
"loss": 0.108,
|
| 2202 |
+
"step": 311
|
| 2203 |
+
},
|
| 2204 |
+
{
|
| 2205 |
+
"epoch": 2.567901234567901,
|
| 2206 |
+
"grad_norm": 0.10969046879996797,
|
| 2207 |
+
"learning_rate": 1.1836281747912125e-05,
|
| 2208 |
+
"loss": 0.1231,
|
| 2209 |
+
"step": 312
|
| 2210 |
+
},
|
| 2211 |
+
{
|
| 2212 |
+
"epoch": 2.5761316872427984,
|
| 2213 |
+
"grad_norm": 0.09910058926653355,
|
| 2214 |
+
"learning_rate": 1.1385595722199438e-05,
|
| 2215 |
+
"loss": 0.1132,
|
| 2216 |
+
"step": 313
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 2.5843621399176957,
|
| 2220 |
+
"grad_norm": 0.0973576033135528,
|
| 2221 |
+
"learning_rate": 1.0943139045973549e-05,
|
| 2222 |
+
"loss": 0.1152,
|
| 2223 |
+
"step": 314
|
| 2224 |
+
},
|
| 2225 |
+
{
|
| 2226 |
+
"epoch": 2.5925925925925926,
|
| 2227 |
+
"grad_norm": 0.10324605327000957,
|
| 2228 |
+
"learning_rate": 1.050895280883668e-05,
|
| 2229 |
+
"loss": 0.119,
|
| 2230 |
+
"step": 315
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 2.60082304526749,
|
| 2234 |
+
"grad_norm": 0.09940459072200475,
|
| 2235 |
+
"learning_rate": 1.0083077332340562e-05,
|
| 2236 |
+
"loss": 0.1086,
|
| 2237 |
+
"step": 316
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 2.6090534979423867,
|
| 2241 |
+
"grad_norm": 0.10054320644748665,
|
| 2242 |
+
"learning_rate": 9.665552166241964e-06,
|
| 2243 |
+
"loss": 0.1149,
|
| 2244 |
+
"step": 317
|
| 2245 |
+
},
|
| 2246 |
+
{
|
| 2247 |
+
"epoch": 2.617283950617284,
|
| 2248 |
+
"grad_norm": 0.10479612628242368,
|
| 2249 |
+
"learning_rate": 9.256416084829778e-06,
|
| 2250 |
+
"loss": 0.1274,
|
| 2251 |
+
"step": 318
|
| 2252 |
+
},
|
| 2253 |
+
{
|
| 2254 |
+
"epoch": 2.625514403292181,
|
| 2255 |
+
"grad_norm": 0.09669311712494659,
|
| 2256 |
+
"learning_rate": 8.855707083324183e-06,
|
| 2257 |
+
"loss": 0.1165,
|
| 2258 |
+
"step": 319
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 2.633744855967078,
|
| 2262 |
+
"grad_norm": 0.09887270600607351,
|
| 2263 |
+
"learning_rate": 8.46346237434813e-06,
|
| 2264 |
+
"loss": 0.1056,
|
| 2265 |
+
"step": 320
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 2.6419753086419755,
|
| 2269 |
+
"grad_norm": 0.10455687155682203,
|
| 2270 |
+
"learning_rate": 8.079718384471557e-06,
|
| 2271 |
+
"loss": 0.1152,
|
| 2272 |
+
"step": 321
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 2.6502057613168724,
|
| 2276 |
+
"grad_norm": 0.10011518680387445,
|
| 2277 |
+
"learning_rate": 7.704510750828542e-06,
|
| 2278 |
+
"loss": 0.1056,
|
| 2279 |
+
"step": 322
|
| 2280 |
+
},
|
| 2281 |
+
{
|
| 2282 |
+
"epoch": 2.6584362139917697,
|
| 2283 |
+
"grad_norm": 0.09682701363965382,
|
| 2284 |
+
"learning_rate": 7.337874317807802e-06,
|
| 2285 |
+
"loss": 0.1034,
|
| 2286 |
+
"step": 323
|
| 2287 |
+
},
|
| 2288 |
+
{
|
| 2289 |
+
"epoch": 2.6666666666666665,
|
| 2290 |
+
"grad_norm": 0.10000970149912758,
|
| 2291 |
+
"learning_rate": 6.979843133816743e-06,
|
| 2292 |
+
"loss": 0.1119,
|
| 2293 |
+
"step": 324
|
| 2294 |
+
},
|
| 2295 |
+
{
|
| 2296 |
+
"epoch": 2.674897119341564,
|
| 2297 |
+
"grad_norm": 0.10465713299866858,
|
| 2298 |
+
"learning_rate": 6.630450448119618e-06,
|
| 2299 |
+
"loss": 0.1183,
|
| 2300 |
+
"step": 325
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 2.6831275720164607,
|
| 2304 |
+
"grad_norm": 0.10068814150000575,
|
| 2305 |
+
"learning_rate": 6.289728707749609e-06,
|
| 2306 |
+
"loss": 0.11,
|
| 2307 |
+
"step": 326
|
| 2308 |
+
},
|
| 2309 |
+
{
|
| 2310 |
+
"epoch": 2.691358024691358,
|
| 2311 |
+
"grad_norm": 0.09310993381960417,
|
| 2312 |
+
"learning_rate": 5.957709554495683e-06,
|
| 2313 |
+
"loss": 0.103,
|
| 2314 |
+
"step": 327
|
| 2315 |
+
},
|
| 2316 |
+
{
|
| 2317 |
+
"epoch": 2.6995884773662553,
|
| 2318 |
+
"grad_norm": 0.10472398535370353,
|
| 2319 |
+
"learning_rate": 5.634423821964074e-06,
|
| 2320 |
+
"loss": 0.1161,
|
| 2321 |
+
"step": 328
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 2.707818930041152,
|
| 2325 |
+
"grad_norm": 0.10645507272996577,
|
| 2326 |
+
"learning_rate": 5.319901532714877e-06,
|
| 2327 |
+
"loss": 0.1217,
|
| 2328 |
+
"step": 329
|
| 2329 |
+
},
|
| 2330 |
+
{
|
| 2331 |
+
"epoch": 2.7160493827160495,
|
| 2332 |
+
"grad_norm": 0.10602188133639034,
|
| 2333 |
+
"learning_rate": 5.014171895473929e-06,
|
| 2334 |
+
"loss": 0.1144,
|
| 2335 |
+
"step": 330
|
| 2336 |
+
},
|
| 2337 |
+
{
|
| 2338 |
+
"epoch": 2.7242798353909468,
|
| 2339 |
+
"grad_norm": 0.10289659614865782,
|
| 2340 |
+
"learning_rate": 4.717263302420283e-06,
|
| 2341 |
+
"loss": 0.1079,
|
| 2342 |
+
"step": 331
|
| 2343 |
+
},
|
| 2344 |
+
{
|
| 2345 |
+
"epoch": 2.7325102880658436,
|
| 2346 |
+
"grad_norm": 0.09777304769239446,
|
| 2347 |
+
"learning_rate": 4.429203326549525e-06,
|
| 2348 |
+
"loss": 0.0999,
|
| 2349 |
+
"step": 332
|
| 2350 |
+
},
|
| 2351 |
+
{
|
| 2352 |
+
"epoch": 2.7407407407407405,
|
| 2353 |
+
"grad_norm": 0.1006102230121861,
|
| 2354 |
+
"learning_rate": 4.1500187191131466e-06,
|
| 2355 |
+
"loss": 0.099,
|
| 2356 |
+
"step": 333
|
| 2357 |
+
},
|
| 2358 |
+
{
|
| 2359 |
+
"epoch": 2.748971193415638,
|
| 2360 |
+
"grad_norm": 0.10180225031662725,
|
| 2361 |
+
"learning_rate": 3.879735407134244e-06,
|
| 2362 |
+
"loss": 0.1173,
|
| 2363 |
+
"step": 334
|
| 2364 |
+
},
|
| 2365 |
+
{
|
| 2366 |
+
"epoch": 2.757201646090535,
|
| 2367 |
+
"grad_norm": 0.10250402250260239,
|
| 2368 |
+
"learning_rate": 3.6183784909997187e-06,
|
| 2369 |
+
"loss": 0.1139,
|
| 2370 |
+
"step": 335
|
| 2371 |
+
},
|
| 2372 |
+
{
|
| 2373 |
+
"epoch": 2.765432098765432,
|
| 2374 |
+
"grad_norm": 0.10615733085722019,
|
| 2375 |
+
"learning_rate": 3.3659722421293783e-06,
|
| 2376 |
+
"loss": 0.1133,
|
| 2377 |
+
"step": 336
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 2.7736625514403292,
|
| 2381 |
+
"grad_norm": 0.10439855178259218,
|
| 2382 |
+
"learning_rate": 3.1225401007217936e-06,
|
| 2383 |
+
"loss": 0.1119,
|
| 2384 |
+
"step": 337
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 2.7818930041152266,
|
| 2388 |
+
"grad_norm": 0.10526144305704047,
|
| 2389 |
+
"learning_rate": 2.8881046735775742e-06,
|
| 2390 |
+
"loss": 0.1219,
|
| 2391 |
+
"step": 338
|
| 2392 |
+
},
|
| 2393 |
+
{
|
| 2394 |
+
"epoch": 2.7901234567901234,
|
| 2395 |
+
"grad_norm": 0.10521105916873999,
|
| 2396 |
+
"learning_rate": 2.66268773199988e-06,
|
| 2397 |
+
"loss": 0.1074,
|
| 2398 |
+
"step": 339
|
| 2399 |
+
},
|
| 2400 |
+
{
|
| 2401 |
+
"epoch": 2.7983539094650207,
|
| 2402 |
+
"grad_norm": 0.10748613106096833,
|
| 2403 |
+
"learning_rate": 2.446310209772684e-06,
|
| 2404 |
+
"loss": 0.1128,
|
| 2405 |
+
"step": 340
|
| 2406 |
+
},
|
| 2407 |
+
{
|
| 2408 |
+
"epoch": 2.8065843621399176,
|
| 2409 |
+
"grad_norm": 0.09914523804031186,
|
| 2410 |
+
"learning_rate": 2.2389922012165944e-06,
|
| 2411 |
+
"loss": 0.1121,
|
| 2412 |
+
"step": 341
|
| 2413 |
+
},
|
| 2414 |
+
{
|
| 2415 |
+
"epoch": 2.814814814814815,
|
| 2416 |
+
"grad_norm": 0.09936886240495578,
|
| 2417 |
+
"learning_rate": 2.0407529593228116e-06,
|
| 2418 |
+
"loss": 0.108,
|
| 2419 |
+
"step": 342
|
| 2420 |
+
},
|
| 2421 |
+
{
|
| 2422 |
+
"epoch": 2.8230452674897117,
|
| 2423 |
+
"grad_norm": 0.09150083914451453,
|
| 2424 |
+
"learning_rate": 1.8516108939651945e-06,
|
| 2425 |
+
"loss": 0.1033,
|
| 2426 |
+
"step": 343
|
| 2427 |
+
},
|
| 2428 |
+
{
|
| 2429 |
+
"epoch": 2.831275720164609,
|
| 2430 |
+
"grad_norm": 0.10108462806007419,
|
| 2431 |
+
"learning_rate": 1.6715835701905603e-06,
|
| 2432 |
+
"loss": 0.1126,
|
| 2433 |
+
"step": 344
|
| 2434 |
+
},
|
| 2435 |
+
{
|
| 2436 |
+
"epoch": 2.8395061728395063,
|
| 2437 |
+
"grad_norm": 0.09387291972239033,
|
| 2438 |
+
"learning_rate": 1.5006877065874336e-06,
|
| 2439 |
+
"loss": 0.1033,
|
| 2440 |
+
"step": 345
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 2.847736625514403,
|
| 2444 |
+
"grad_norm": 0.10136947076698849,
|
| 2445 |
+
"learning_rate": 1.3389391737335112e-06,
|
| 2446 |
+
"loss": 0.1104,
|
| 2447 |
+
"step": 346
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 2.8559670781893005,
|
| 2451 |
+
"grad_norm": 0.09974875048963104,
|
| 2452 |
+
"learning_rate": 1.1863529927217732e-06,
|
| 2453 |
+
"loss": 0.1043,
|
| 2454 |
+
"step": 347
|
| 2455 |
+
},
|
| 2456 |
+
{
|
| 2457 |
+
"epoch": 2.8641975308641974,
|
| 2458 |
+
"grad_norm": 0.1025752381453182,
|
| 2459 |
+
"learning_rate": 1.0429433337655115e-06,
|
| 2460 |
+
"loss": 0.1122,
|
| 2461 |
+
"step": 348
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 2.8724279835390947,
|
| 2465 |
+
"grad_norm": 0.10454016619792661,
|
| 2466 |
+
"learning_rate": 9.087235148824368e-07,
|
| 2467 |
+
"loss": 0.1184,
|
| 2468 |
+
"step": 349
|
| 2469 |
+
},
|
| 2470 |
+
{
|
| 2471 |
+
"epoch": 2.8806584362139915,
|
| 2472 |
+
"grad_norm": 0.09589319596591109,
|
| 2473 |
+
"learning_rate": 7.837060006577801e-07,
|
| 2474 |
+
"loss": 0.1046,
|
| 2475 |
+
"step": 350
|
| 2476 |
+
},
|
| 2477 |
+
{
|
| 2478 |
+
"epoch": 2.888888888888889,
|
| 2479 |
+
"grad_norm": 0.09451088377127861,
|
| 2480 |
+
"learning_rate": 6.679024010868618e-07,
|
| 2481 |
+
"loss": 0.1103,
|
| 2482 |
+
"step": 351
|
| 2483 |
+
},
|
| 2484 |
+
{
|
| 2485 |
+
"epoch": 2.897119341563786,
|
| 2486 |
+
"grad_norm": 0.09770471400992828,
|
| 2487 |
+
"learning_rate": 5.613234704967996e-07,
|
| 2488 |
+
"loss": 0.1027,
|
| 2489 |
+
"step": 352
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 2.905349794238683,
|
| 2493 |
+
"grad_norm": 0.09543457833989027,
|
| 2494 |
+
"learning_rate": 4.639791065478738e-07,
|
| 2495 |
+
"loss": 0.1025,
|
| 2496 |
+
"step": 353
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 2.9135802469135803,
|
| 2500 |
+
"grad_norm": 0.10382486799240734,
|
| 2501 |
+
"learning_rate": 3.758783493142737e-07,
|
| 2502 |
+
"loss": 0.118,
|
| 2503 |
+
"step": 354
|
| 2504 |
+
},
|
| 2505 |
+
{
|
| 2506 |
+
"epoch": 2.9218106995884776,
|
| 2507 |
+
"grad_norm": 0.1030393593220336,
|
| 2508 |
+
"learning_rate": 2.9702938044468e-07,
|
| 2509 |
+
"loss": 0.1179,
|
| 2510 |
+
"step": 355
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 2.9300411522633745,
|
| 2514 |
+
"grad_norm": 0.10073287500645238,
|
| 2515 |
+
"learning_rate": 2.2743952240236176e-07,
|
| 2516 |
+
"loss": 0.1058,
|
| 2517 |
+
"step": 356
|
| 2518 |
+
},
|
| 2519 |
+
{
|
| 2520 |
+
"epoch": 2.9382716049382713,
|
| 2521 |
+
"grad_norm": 0.10064431575396734,
|
| 2522 |
+
"learning_rate": 1.6711523778520921e-07,
|
| 2523 |
+
"loss": 0.1061,
|
| 2524 |
+
"step": 357
|
| 2525 |
+
},
|
| 2526 |
+
{
|
| 2527 |
+
"epoch": 2.9465020576131686,
|
| 2528 |
+
"grad_norm": 0.08907660041993512,
|
| 2529 |
+
"learning_rate": 1.1606212872559141e-07,
|
| 2530 |
+
"loss": 0.0985,
|
| 2531 |
+
"step": 358
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 2.954732510288066,
|
| 2535 |
+
"grad_norm": 0.10069703961300915,
|
| 2536 |
+
"learning_rate": 7.428493637002821e-08,
|
| 2537 |
+
"loss": 0.1107,
|
| 2538 |
+
"step": 359
|
| 2539 |
+
},
|
| 2540 |
+
{
|
| 2541 |
+
"epoch": 2.962962962962963,
|
| 2542 |
+
"grad_norm": 0.09834578070623724,
|
| 2543 |
+
"learning_rate": 4.178754043898669e-08,
|
| 2544 |
+
"loss": 0.1056,
|
| 2545 |
+
"step": 360
|
| 2546 |
+
},
|
| 2547 |
+
{
|
| 2548 |
+
"epoch": 2.97119341563786,
|
| 2549 |
+
"grad_norm": 0.10056360393735715,
|
| 2550 |
+
"learning_rate": 1.8572958866514e-08,
|
| 2551 |
+
"loss": 0.1123,
|
| 2552 |
+
"step": 361
|
| 2553 |
+
},
|
| 2554 |
+
{
|
| 2555 |
+
"epoch": 2.9794238683127574,
|
| 2556 |
+
"grad_norm": 0.0943621498377698,
|
| 2557 |
+
"learning_rate": 4.643347520005836e-09,
|
| 2558 |
+
"loss": 0.1045,
|
| 2559 |
+
"step": 362
|
| 2560 |
+
},
|
| 2561 |
+
{
|
| 2562 |
+
"epoch": 2.9876543209876543,
|
| 2563 |
+
"grad_norm": 0.09983652481727213,
|
| 2564 |
+
"learning_rate": 0.0,
|
| 2565 |
+
"loss": 0.1135,
|
| 2566 |
+
"step": 363
|
| 2567 |
+
},
|
| 2568 |
+
{
|
| 2569 |
+
"epoch": 2.9876543209876543,
|
| 2570 |
+
"eval_loss": 0.13842210173606873,
|
| 2571 |
+
"eval_runtime": 21.2699,
|
| 2572 |
+
"eval_samples_per_second": 38.411,
|
| 2573 |
+
"eval_steps_per_second": 1.222,
|
| 2574 |
+
"step": 363
|
| 2575 |
+
},
|
| 2576 |
+
{
|
| 2577 |
+
"epoch": 2.9876543209876543,
|
| 2578 |
+
"step": 363,
|
| 2579 |
+
"total_flos": 9.618537139981517e+16,
|
| 2580 |
+
"train_loss": 0.21143763487742953,
|
| 2581 |
+
"train_runtime": 3172.33,
|
| 2582 |
+
"train_samples_per_second": 14.668,
|
| 2583 |
+
"train_steps_per_second": 0.114
|
| 2584 |
+
}
|
| 2585 |
+
],
|
| 2586 |
+
"logging_steps": 1,
|
| 2587 |
+
"max_steps": 363,
|
| 2588 |
+
"num_input_tokens_seen": 0,
|
| 2589 |
+
"num_train_epochs": 3,
|
| 2590 |
+
"save_steps": 500,
|
| 2591 |
+
"stateful_callbacks": {
|
| 2592 |
+
"TrainerControl": {
|
| 2593 |
+
"args": {
|
| 2594 |
+
"should_epoch_stop": false,
|
| 2595 |
+
"should_evaluate": false,
|
| 2596 |
+
"should_log": false,
|
| 2597 |
+
"should_save": true,
|
| 2598 |
+
"should_training_stop": true
|
| 2599 |
+
},
|
| 2600 |
+
"attributes": {}
|
| 2601 |
+
}
|
| 2602 |
+
},
|
| 2603 |
+
"total_flos": 9.618537139981517e+16,
|
| 2604 |
+
"train_batch_size": 8,
|
| 2605 |
+
"trial_name": null,
|
| 2606 |
+
"trial_params": null
|
| 2607 |
+
}
|