Model save
Browse files- README.md +85 -0
- all_results.json +9 -0
- generation_config.json +13 -0
- train_results.json +9 -0
- trainer_state.json +1609 -0
README.md
ADDED
|
@@ -0,0 +1,85 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: llama3.2
|
| 4 |
+
base_model: tanliboy/llama-3.2-3b
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: llama-3.2-3b-sft-2
|
| 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 |
+
# llama-3.2-3b-sft-2
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [tanliboy/llama-3.2-3b](https://huggingface.co/tanliboy/llama-3.2-3b) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.6744
|
| 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: 5e-06
|
| 41 |
+
- train_batch_size: 8
|
| 42 |
+
- eval_batch_size: 8
|
| 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: 64
|
| 49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 50 |
+
- lr_scheduler_type: cosine
|
| 51 |
+
- lr_scheduler_warmup_steps: 1000
|
| 52 |
+
- training_steps: 10000
|
| 53 |
+
|
| 54 |
+
### Training results
|
| 55 |
+
|
| 56 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 57 |
+
|:-------------:|:------:|:-----:|:---------------:|
|
| 58 |
+
| 0.7792 | 0.0673 | 500 | 0.7726 |
|
| 59 |
+
| 0.7496 | 0.1345 | 1000 | 0.7444 |
|
| 60 |
+
| 0.7243 | 0.2018 | 1500 | 0.7296 |
|
| 61 |
+
| 0.7178 | 0.2691 | 2000 | 0.7197 |
|
| 62 |
+
| 0.7077 | 0.3363 | 2500 | 0.7127 |
|
| 63 |
+
| 0.6992 | 0.4036 | 3000 | 0.7066 |
|
| 64 |
+
| 0.6992 | 0.4708 | 3500 | 0.7012 |
|
| 65 |
+
| 0.6945 | 0.5381 | 4000 | 0.6965 |
|
| 66 |
+
| 0.6879 | 0.6054 | 4500 | 0.6920 |
|
| 67 |
+
| 0.6901 | 0.6726 | 5000 | 0.6879 |
|
| 68 |
+
| 0.6759 | 0.7399 | 5500 | 0.6844 |
|
| 69 |
+
| 0.6752 | 0.8072 | 6000 | 0.6812 |
|
| 70 |
+
| 0.6826 | 0.8744 | 6500 | 0.6783 |
|
| 71 |
+
| 0.6804 | 0.9417 | 7000 | 0.6758 |
|
| 72 |
+
| 0.6131 | 1.0089 | 7500 | 0.6764 |
|
| 73 |
+
| 0.6012 | 1.0762 | 8000 | 0.6758 |
|
| 74 |
+
| 0.6136 | 1.1435 | 8500 | 0.6751 |
|
| 75 |
+
| 0.6127 | 1.2107 | 9000 | 0.6747 |
|
| 76 |
+
| 0.6076 | 1.2780 | 9500 | 0.6745 |
|
| 77 |
+
| 0.6033 | 1.3453 | 10000 | 0.6744 |
|
| 78 |
+
|
| 79 |
+
|
| 80 |
+
### Framework versions
|
| 81 |
+
|
| 82 |
+
- Transformers 4.44.2
|
| 83 |
+
- Pytorch 2.4.0+cu121
|
| 84 |
+
- Datasets 2.19.1
|
| 85 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.3452613170108294,
|
| 3 |
+
"total_flos": 1119551134941184.0,
|
| 4 |
+
"train_loss": 0.6844510423541069,
|
| 5 |
+
"train_runtime": 79339.5992,
|
| 6 |
+
"train_samples": 951461,
|
| 7 |
+
"train_samples_per_second": 16.133,
|
| 8 |
+
"train_steps_per_second": 0.126
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 128000,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
128001,
|
| 7 |
+
128008,
|
| 8 |
+
128009
|
| 9 |
+
],
|
| 10 |
+
"temperature": 0.6,
|
| 11 |
+
"top_p": 0.9,
|
| 12 |
+
"transformers_version": "4.44.2"
|
| 13 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.3452613170108294,
|
| 3 |
+
"total_flos": 1119551134941184.0,
|
| 4 |
+
"train_loss": 0.6844510423541069,
|
| 5 |
+
"train_runtime": 79339.5992,
|
| 6 |
+
"train_samples": 951461,
|
| 7 |
+
"train_samples_per_second": 16.133,
|
| 8 |
+
"train_steps_per_second": 0.126
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1609 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6744288802146912,
|
| 3 |
+
"best_model_checkpoint": "models/llama-3.2-3b-sft-2/checkpoint-10000",
|
| 4 |
+
"epoch": 1.3452613170108294,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 10000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.00013452613170108295,
|
| 13 |
+
"grad_norm": 5.089283466339111,
|
| 14 |
+
"learning_rate": 5e-09,
|
| 15 |
+
"loss": 1.0115,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.006726306585054147,
|
| 20 |
+
"grad_norm": 3.6531453132629395,
|
| 21 |
+
"learning_rate": 2.5000000000000004e-07,
|
| 22 |
+
"loss": 1.0329,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.013452613170108294,
|
| 27 |
+
"grad_norm": 1.954504132270813,
|
| 28 |
+
"learning_rate": 5.000000000000001e-07,
|
| 29 |
+
"loss": 0.9628,
|
| 30 |
+
"step": 100
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.02017891975516244,
|
| 34 |
+
"grad_norm": 1.6202716827392578,
|
| 35 |
+
"learning_rate": 7.5e-07,
|
| 36 |
+
"loss": 0.882,
|
| 37 |
+
"step": 150
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.02690522634021659,
|
| 41 |
+
"grad_norm": 1.5833114385604858,
|
| 42 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 43 |
+
"loss": 0.8489,
|
| 44 |
+
"step": 200
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.033631532925270734,
|
| 48 |
+
"grad_norm": 1.6729789972305298,
|
| 49 |
+
"learning_rate": 1.25e-06,
|
| 50 |
+
"loss": 0.8464,
|
| 51 |
+
"step": 250
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.04035783951032488,
|
| 55 |
+
"grad_norm": 1.551665186882019,
|
| 56 |
+
"learning_rate": 1.5e-06,
|
| 57 |
+
"loss": 0.8205,
|
| 58 |
+
"step": 300
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.047084146095379024,
|
| 62 |
+
"grad_norm": 1.6382861137390137,
|
| 63 |
+
"learning_rate": 1.75e-06,
|
| 64 |
+
"loss": 0.7994,
|
| 65 |
+
"step": 350
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.05381045268043318,
|
| 69 |
+
"grad_norm": 1.8948196172714233,
|
| 70 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 71 |
+
"loss": 0.7942,
|
| 72 |
+
"step": 400
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.06053675926548732,
|
| 76 |
+
"grad_norm": 1.5274723768234253,
|
| 77 |
+
"learning_rate": 2.25e-06,
|
| 78 |
+
"loss": 0.7843,
|
| 79 |
+
"step": 450
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.06726306585054147,
|
| 83 |
+
"grad_norm": 2.083868980407715,
|
| 84 |
+
"learning_rate": 2.5e-06,
|
| 85 |
+
"loss": 0.7792,
|
| 86 |
+
"step": 500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.06726306585054147,
|
| 90 |
+
"eval_loss": 0.772584855556488,
|
| 91 |
+
"eval_runtime": 412.1529,
|
| 92 |
+
"eval_samples_per_second": 121.501,
|
| 93 |
+
"eval_steps_per_second": 1.9,
|
| 94 |
+
"step": 500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.07398937243559561,
|
| 98 |
+
"grad_norm": 1.4476754665374756,
|
| 99 |
+
"learning_rate": 2.7500000000000004e-06,
|
| 100 |
+
"loss": 0.7598,
|
| 101 |
+
"step": 550
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.08071567902064976,
|
| 105 |
+
"grad_norm": 1.5177018642425537,
|
| 106 |
+
"learning_rate": 3e-06,
|
| 107 |
+
"loss": 0.7672,
|
| 108 |
+
"step": 600
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.0874419856057039,
|
| 112 |
+
"grad_norm": 1.5025829076766968,
|
| 113 |
+
"learning_rate": 3.2500000000000002e-06,
|
| 114 |
+
"loss": 0.7442,
|
| 115 |
+
"step": 650
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.09416829219075805,
|
| 119 |
+
"grad_norm": 1.6595162153244019,
|
| 120 |
+
"learning_rate": 3.5e-06,
|
| 121 |
+
"loss": 0.7562,
|
| 122 |
+
"step": 700
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.10089459877581221,
|
| 126 |
+
"grad_norm": 1.4664040803909302,
|
| 127 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 128 |
+
"loss": 0.7485,
|
| 129 |
+
"step": 750
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.10762090536086635,
|
| 133 |
+
"grad_norm": 1.5404261350631714,
|
| 134 |
+
"learning_rate": 4.000000000000001e-06,
|
| 135 |
+
"loss": 0.7482,
|
| 136 |
+
"step": 800
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.1143472119459205,
|
| 140 |
+
"grad_norm": 1.553759217262268,
|
| 141 |
+
"learning_rate": 4.25e-06,
|
| 142 |
+
"loss": 0.7475,
|
| 143 |
+
"step": 850
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.12107351853097464,
|
| 147 |
+
"grad_norm": 1.5157835483551025,
|
| 148 |
+
"learning_rate": 4.5e-06,
|
| 149 |
+
"loss": 0.7403,
|
| 150 |
+
"step": 900
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.12779982511602878,
|
| 154 |
+
"grad_norm": 1.4276628494262695,
|
| 155 |
+
"learning_rate": 4.75e-06,
|
| 156 |
+
"loss": 0.7392,
|
| 157 |
+
"step": 950
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.13452613170108293,
|
| 161 |
+
"grad_norm": 1.5712127685546875,
|
| 162 |
+
"learning_rate": 5e-06,
|
| 163 |
+
"loss": 0.7496,
|
| 164 |
+
"step": 1000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.13452613170108293,
|
| 168 |
+
"eval_loss": 0.7444220185279846,
|
| 169 |
+
"eval_runtime": 412.2013,
|
| 170 |
+
"eval_samples_per_second": 121.487,
|
| 171 |
+
"eval_steps_per_second": 1.9,
|
| 172 |
+
"step": 1000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.1412524382861371,
|
| 176 |
+
"grad_norm": 1.3969111442565918,
|
| 177 |
+
"learning_rate": 4.9996192378909785e-06,
|
| 178 |
+
"loss": 0.7548,
|
| 179 |
+
"step": 1050
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.14797874487119123,
|
| 183 |
+
"grad_norm": 1.4484981298446655,
|
| 184 |
+
"learning_rate": 4.99847706754774e-06,
|
| 185 |
+
"loss": 0.73,
|
| 186 |
+
"step": 1100
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.15470505145624538,
|
| 190 |
+
"grad_norm": 1.4184221029281616,
|
| 191 |
+
"learning_rate": 4.9965738368864345e-06,
|
| 192 |
+
"loss": 0.734,
|
| 193 |
+
"step": 1150
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.16143135804129952,
|
| 197 |
+
"grad_norm": 1.4049128293991089,
|
| 198 |
+
"learning_rate": 4.993910125649561e-06,
|
| 199 |
+
"loss": 0.7295,
|
| 200 |
+
"step": 1200
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.16815766462635368,
|
| 204 |
+
"grad_norm": 1.3905514478683472,
|
| 205 |
+
"learning_rate": 4.990486745229364e-06,
|
| 206 |
+
"loss": 0.7405,
|
| 207 |
+
"step": 1250
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.1748839712114078,
|
| 211 |
+
"grad_norm": 1.5445802211761475,
|
| 212 |
+
"learning_rate": 4.986304738420684e-06,
|
| 213 |
+
"loss": 0.7198,
|
| 214 |
+
"step": 1300
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.18161027779646197,
|
| 218 |
+
"grad_norm": 1.4158939123153687,
|
| 219 |
+
"learning_rate": 4.981365379103306e-06,
|
| 220 |
+
"loss": 0.7355,
|
| 221 |
+
"step": 1350
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.1883365843815161,
|
| 225 |
+
"grad_norm": 1.3896397352218628,
|
| 226 |
+
"learning_rate": 4.975670171853926e-06,
|
| 227 |
+
"loss": 0.7168,
|
| 228 |
+
"step": 1400
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.19506289096657026,
|
| 232 |
+
"grad_norm": 1.5642178058624268,
|
| 233 |
+
"learning_rate": 4.9692208514878445e-06,
|
| 234 |
+
"loss": 0.7369,
|
| 235 |
+
"step": 1450
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.20178919755162442,
|
| 239 |
+
"grad_norm": 1.287514328956604,
|
| 240 |
+
"learning_rate": 4.962019382530521e-06,
|
| 241 |
+
"loss": 0.7243,
|
| 242 |
+
"step": 1500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.20178919755162442,
|
| 246 |
+
"eval_loss": 0.7295696139335632,
|
| 247 |
+
"eval_runtime": 408.0925,
|
| 248 |
+
"eval_samples_per_second": 122.71,
|
| 249 |
+
"eval_steps_per_second": 1.919,
|
| 250 |
+
"step": 1500
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.20851550413667855,
|
| 254 |
+
"grad_norm": 1.380998969078064,
|
| 255 |
+
"learning_rate": 4.9540679586191605e-06,
|
| 256 |
+
"loss": 0.7309,
|
| 257 |
+
"step": 1550
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.2152418107217327,
|
| 261 |
+
"grad_norm": 1.4188984632492065,
|
| 262 |
+
"learning_rate": 4.9453690018345144e-06,
|
| 263 |
+
"loss": 0.7217,
|
| 264 |
+
"step": 1600
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.22196811730678684,
|
| 268 |
+
"grad_norm": 1.306766390800476,
|
| 269 |
+
"learning_rate": 4.935925161963089e-06,
|
| 270 |
+
"loss": 0.7336,
|
| 271 |
+
"step": 1650
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.228694423891841,
|
| 275 |
+
"grad_norm": 1.460458517074585,
|
| 276 |
+
"learning_rate": 4.925739315689991e-06,
|
| 277 |
+
"loss": 0.7258,
|
| 278 |
+
"step": 1700
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.23542073047689513,
|
| 282 |
+
"grad_norm": 1.4665378332138062,
|
| 283 |
+
"learning_rate": 4.914814565722671e-06,
|
| 284 |
+
"loss": 0.7249,
|
| 285 |
+
"step": 1750
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.2421470370619493,
|
| 289 |
+
"grad_norm": 1.5657004117965698,
|
| 290 |
+
"learning_rate": 4.903154239845798e-06,
|
| 291 |
+
"loss": 0.7253,
|
| 292 |
+
"step": 1800
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.24887334364700342,
|
| 296 |
+
"grad_norm": 1.3820271492004395,
|
| 297 |
+
"learning_rate": 4.890761889907589e-06,
|
| 298 |
+
"loss": 0.7151,
|
| 299 |
+
"step": 1850
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.25559965023205755,
|
| 303 |
+
"grad_norm": 1.4721177816390991,
|
| 304 |
+
"learning_rate": 4.8776412907378845e-06,
|
| 305 |
+
"loss": 0.7251,
|
| 306 |
+
"step": 1900
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.26232595681711174,
|
| 310 |
+
"grad_norm": 1.4119197130203247,
|
| 311 |
+
"learning_rate": 4.863796438998293e-06,
|
| 312 |
+
"loss": 0.7188,
|
| 313 |
+
"step": 1950
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.26905226340216587,
|
| 317 |
+
"grad_norm": 1.3871151208877563,
|
| 318 |
+
"learning_rate": 4.849231551964771e-06,
|
| 319 |
+
"loss": 0.7178,
|
| 320 |
+
"step": 2000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.26905226340216587,
|
| 324 |
+
"eval_loss": 0.7197288870811462,
|
| 325 |
+
"eval_runtime": 407.583,
|
| 326 |
+
"eval_samples_per_second": 122.863,
|
| 327 |
+
"eval_steps_per_second": 1.921,
|
| 328 |
+
"step": 2000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.27577856998722,
|
| 332 |
+
"grad_norm": 1.4082683324813843,
|
| 333 |
+
"learning_rate": 4.833951066243004e-06,
|
| 334 |
+
"loss": 0.7255,
|
| 335 |
+
"step": 2050
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 0.2825048765722742,
|
| 339 |
+
"grad_norm": 1.3477662801742554,
|
| 340 |
+
"learning_rate": 4.817959636416969e-06,
|
| 341 |
+
"loss": 0.7141,
|
| 342 |
+
"step": 2100
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 0.2892311831573283,
|
| 346 |
+
"grad_norm": 1.316319227218628,
|
| 347 |
+
"learning_rate": 4.801262133631101e-06,
|
| 348 |
+
"loss": 0.7146,
|
| 349 |
+
"step": 2150
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 0.29595748974238245,
|
| 353 |
+
"grad_norm": 1.3264553546905518,
|
| 354 |
+
"learning_rate": 4.783863644106502e-06,
|
| 355 |
+
"loss": 0.7176,
|
| 356 |
+
"step": 2200
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.3026837963274366,
|
| 360 |
+
"grad_norm": 1.4020483493804932,
|
| 361 |
+
"learning_rate": 4.765769467591626e-06,
|
| 362 |
+
"loss": 0.7082,
|
| 363 |
+
"step": 2250
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.30941010291249077,
|
| 367 |
+
"grad_norm": 1.4155000448226929,
|
| 368 |
+
"learning_rate": 4.746985115747918e-06,
|
| 369 |
+
"loss": 0.7133,
|
| 370 |
+
"step": 2300
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.3161364094975449,
|
| 374 |
+
"grad_norm": 1.3636876344680786,
|
| 375 |
+
"learning_rate": 4.72751631047092e-06,
|
| 376 |
+
"loss": 0.721,
|
| 377 |
+
"step": 2350
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.32286271608259903,
|
| 381 |
+
"grad_norm": 1.3558250665664673,
|
| 382 |
+
"learning_rate": 4.707368982147318e-06,
|
| 383 |
+
"loss": 0.7056,
|
| 384 |
+
"step": 2400
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.32958902266765316,
|
| 388 |
+
"grad_norm": 1.3763400316238403,
|
| 389 |
+
"learning_rate": 4.68654926784849e-06,
|
| 390 |
+
"loss": 0.7157,
|
| 391 |
+
"step": 2450
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.33631532925270735,
|
| 395 |
+
"grad_norm": 1.4673148393630981,
|
| 396 |
+
"learning_rate": 4.665063509461098e-06,
|
| 397 |
+
"loss": 0.7077,
|
| 398 |
+
"step": 2500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.33631532925270735,
|
| 402 |
+
"eval_loss": 0.7126947045326233,
|
| 403 |
+
"eval_runtime": 407.7534,
|
| 404 |
+
"eval_samples_per_second": 122.812,
|
| 405 |
+
"eval_steps_per_second": 1.92,
|
| 406 |
+
"step": 2500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.3430416358377615,
|
| 410 |
+
"grad_norm": 1.5427167415618896,
|
| 411 |
+
"learning_rate": 4.642918251755281e-06,
|
| 412 |
+
"loss": 0.7216,
|
| 413 |
+
"step": 2550
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.3497679424228156,
|
| 417 |
+
"grad_norm": 1.4927399158477783,
|
| 418 |
+
"learning_rate": 4.620120240391065e-06,
|
| 419 |
+
"loss": 0.7096,
|
| 420 |
+
"step": 2600
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.3564942490078698,
|
| 424 |
+
"grad_norm": 1.3618338108062744,
|
| 425 |
+
"learning_rate": 4.596676419863561e-06,
|
| 426 |
+
"loss": 0.7106,
|
| 427 |
+
"step": 2650
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.36322055559292393,
|
| 431 |
+
"grad_norm": 1.3076339960098267,
|
| 432 |
+
"learning_rate": 4.572593931387604e-06,
|
| 433 |
+
"loss": 0.7213,
|
| 434 |
+
"step": 2700
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.36994686217797806,
|
| 438 |
+
"grad_norm": 1.2195158004760742,
|
| 439 |
+
"learning_rate": 4.54788011072248e-06,
|
| 440 |
+
"loss": 0.7102,
|
| 441 |
+
"step": 2750
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.3766731687630322,
|
| 445 |
+
"grad_norm": 1.341899037361145,
|
| 446 |
+
"learning_rate": 4.522542485937369e-06,
|
| 447 |
+
"loss": 0.7021,
|
| 448 |
+
"step": 2800
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.3833994753480864,
|
| 452 |
+
"grad_norm": 1.4170920848846436,
|
| 453 |
+
"learning_rate": 4.496588775118232e-06,
|
| 454 |
+
"loss": 0.6942,
|
| 455 |
+
"step": 2850
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 0.3901257819331405,
|
| 459 |
+
"grad_norm": 1.502806305885315,
|
| 460 |
+
"learning_rate": 4.470026884016805e-06,
|
| 461 |
+
"loss": 0.6947,
|
| 462 |
+
"step": 2900
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 0.39685208851819465,
|
| 466 |
+
"grad_norm": 1.3311142921447754,
|
| 467 |
+
"learning_rate": 4.442864903642428e-06,
|
| 468 |
+
"loss": 0.6926,
|
| 469 |
+
"step": 2950
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.40357839510324883,
|
| 473 |
+
"grad_norm": 1.3881827592849731,
|
| 474 |
+
"learning_rate": 4.415111107797445e-06,
|
| 475 |
+
"loss": 0.6992,
|
| 476 |
+
"step": 3000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.40357839510324883,
|
| 480 |
+
"eval_loss": 0.7066433429718018,
|
| 481 |
+
"eval_runtime": 407.4989,
|
| 482 |
+
"eval_samples_per_second": 122.889,
|
| 483 |
+
"eval_steps_per_second": 1.921,
|
| 484 |
+
"step": 3000
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 0.41030470168830296,
|
| 488 |
+
"grad_norm": 1.4951732158660889,
|
| 489 |
+
"learning_rate": 4.386773950556931e-06,
|
| 490 |
+
"loss": 0.6967,
|
| 491 |
+
"step": 3050
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.4170310082733571,
|
| 495 |
+
"grad_norm": 1.3532462120056152,
|
| 496 |
+
"learning_rate": 4.357862063693486e-06,
|
| 497 |
+
"loss": 0.7056,
|
| 498 |
+
"step": 3100
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.4237573148584112,
|
| 502 |
+
"grad_norm": 1.3927810192108154,
|
| 503 |
+
"learning_rate": 4.328384254047927e-06,
|
| 504 |
+
"loss": 0.6962,
|
| 505 |
+
"step": 3150
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.4304836214434654,
|
| 509 |
+
"grad_norm": 1.2891749143600464,
|
| 510 |
+
"learning_rate": 4.2983495008466285e-06,
|
| 511 |
+
"loss": 0.7193,
|
| 512 |
+
"step": 3200
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.43720992802851955,
|
| 516 |
+
"grad_norm": 1.3432807922363281,
|
| 517 |
+
"learning_rate": 4.267766952966369e-06,
|
| 518 |
+
"loss": 0.7008,
|
| 519 |
+
"step": 3250
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.4439362346135737,
|
| 523 |
+
"grad_norm": 1.5565185546875,
|
| 524 |
+
"learning_rate": 4.236645926147493e-06,
|
| 525 |
+
"loss": 0.6939,
|
| 526 |
+
"step": 3300
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.4506625411986278,
|
| 530 |
+
"grad_norm": 1.3859912157058716,
|
| 531 |
+
"learning_rate": 4.204995900156247e-06,
|
| 532 |
+
"loss": 0.6953,
|
| 533 |
+
"step": 3350
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 0.457388847783682,
|
| 537 |
+
"grad_norm": 1.358071208000183,
|
| 538 |
+
"learning_rate": 4.172826515897146e-06,
|
| 539 |
+
"loss": 0.6847,
|
| 540 |
+
"step": 3400
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 0.4641151543687361,
|
| 544 |
+
"grad_norm": 1.3290272951126099,
|
| 545 |
+
"learning_rate": 4.140147572476269e-06,
|
| 546 |
+
"loss": 0.7069,
|
| 547 |
+
"step": 3450
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 0.47084146095379026,
|
| 551 |
+
"grad_norm": 1.387556552886963,
|
| 552 |
+
"learning_rate": 4.106969024216348e-06,
|
| 553 |
+
"loss": 0.6992,
|
| 554 |
+
"step": 3500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.47084146095379026,
|
| 558 |
+
"eval_loss": 0.7012136578559875,
|
| 559 |
+
"eval_runtime": 407.5913,
|
| 560 |
+
"eval_samples_per_second": 122.861,
|
| 561 |
+
"eval_steps_per_second": 1.921,
|
| 562 |
+
"step": 3500
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.47756776753884445,
|
| 566 |
+
"grad_norm": 1.4065874814987183,
|
| 567 |
+
"learning_rate": 4.073300977624594e-06,
|
| 568 |
+
"loss": 0.7044,
|
| 569 |
+
"step": 3550
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.4842940741238986,
|
| 573 |
+
"grad_norm": 1.3837950229644775,
|
| 574 |
+
"learning_rate": 4.039153688314146e-06,
|
| 575 |
+
"loss": 0.7002,
|
| 576 |
+
"step": 3600
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.4910203807089527,
|
| 580 |
+
"grad_norm": 1.345460057258606,
|
| 581 |
+
"learning_rate": 4.0045375578801216e-06,
|
| 582 |
+
"loss": 0.692,
|
| 583 |
+
"step": 3650
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.49774668729400684,
|
| 587 |
+
"grad_norm": 1.3991773128509521,
|
| 588 |
+
"learning_rate": 3.969463130731183e-06,
|
| 589 |
+
"loss": 0.691,
|
| 590 |
+
"step": 3700
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.504472993879061,
|
| 594 |
+
"grad_norm": 1.4494290351867676,
|
| 595 |
+
"learning_rate": 3.933941090877615e-06,
|
| 596 |
+
"loss": 0.6975,
|
| 597 |
+
"step": 3750
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.5111993004641151,
|
| 601 |
+
"grad_norm": 1.4690207242965698,
|
| 602 |
+
"learning_rate": 3.897982258676867e-06,
|
| 603 |
+
"loss": 0.6951,
|
| 604 |
+
"step": 3800
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.5179256070491693,
|
| 608 |
+
"grad_norm": 1.2572777271270752,
|
| 609 |
+
"learning_rate": 3.861597587537568e-06,
|
| 610 |
+
"loss": 0.6997,
|
| 611 |
+
"step": 3850
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.5246519136342235,
|
| 615 |
+
"grad_norm": 1.416960597038269,
|
| 616 |
+
"learning_rate": 3.824798160583012e-06,
|
| 617 |
+
"loss": 0.6883,
|
| 618 |
+
"step": 3900
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.5313782202192776,
|
| 622 |
+
"grad_norm": 1.3925755023956299,
|
| 623 |
+
"learning_rate": 3.787595187275136e-06,
|
| 624 |
+
"loss": 0.6894,
|
| 625 |
+
"step": 3950
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.5381045268043317,
|
| 629 |
+
"grad_norm": 1.2715755701065063,
|
| 630 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 631 |
+
"loss": 0.6945,
|
| 632 |
+
"step": 4000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.5381045268043317,
|
| 636 |
+
"eval_loss": 0.6965380907058716,
|
| 637 |
+
"eval_runtime": 407.4593,
|
| 638 |
+
"eval_samples_per_second": 122.901,
|
| 639 |
+
"eval_steps_per_second": 1.922,
|
| 640 |
+
"step": 4000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.5448308333893859,
|
| 644 |
+
"grad_norm": 1.1935741901397705,
|
| 645 |
+
"learning_rate": 3.7120240506158433e-06,
|
| 646 |
+
"loss": 0.6899,
|
| 647 |
+
"step": 4050
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.55155713997444,
|
| 651 |
+
"grad_norm": 1.3253408670425415,
|
| 652 |
+
"learning_rate": 3.6736789069647273e-06,
|
| 653 |
+
"loss": 0.6964,
|
| 654 |
+
"step": 4100
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.5582834465594941,
|
| 658 |
+
"grad_norm": 1.4265397787094116,
|
| 659 |
+
"learning_rate": 3.634976249348867e-06,
|
| 660 |
+
"loss": 0.6867,
|
| 661 |
+
"step": 4150
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.5650097531445484,
|
| 665 |
+
"grad_norm": 1.3048572540283203,
|
| 666 |
+
"learning_rate": 3.595927866972694e-06,
|
| 667 |
+
"loss": 0.6959,
|
| 668 |
+
"step": 4200
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.5717360597296025,
|
| 672 |
+
"grad_norm": 1.495254397392273,
|
| 673 |
+
"learning_rate": 3.556545654351749e-06,
|
| 674 |
+
"loss": 0.6841,
|
| 675 |
+
"step": 4250
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.5784623663146566,
|
| 679 |
+
"grad_norm": 1.3568543195724487,
|
| 680 |
+
"learning_rate": 3.516841607689501e-06,
|
| 681 |
+
"loss": 0.6911,
|
| 682 |
+
"step": 4300
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.5851886728997108,
|
| 686 |
+
"grad_norm": 1.274553656578064,
|
| 687 |
+
"learning_rate": 3.476827821223184e-06,
|
| 688 |
+
"loss": 0.6819,
|
| 689 |
+
"step": 4350
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.5919149794847649,
|
| 693 |
+
"grad_norm": 1.3335059881210327,
|
| 694 |
+
"learning_rate": 3.436516483539781e-06,
|
| 695 |
+
"loss": 0.7018,
|
| 696 |
+
"step": 4400
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.598641286069819,
|
| 700 |
+
"grad_norm": 1.265262246131897,
|
| 701 |
+
"learning_rate": 3.39591987386325e-06,
|
| 702 |
+
"loss": 0.6821,
|
| 703 |
+
"step": 4450
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.6053675926548732,
|
| 707 |
+
"grad_norm": 1.3332948684692383,
|
| 708 |
+
"learning_rate": 3.3550503583141726e-06,
|
| 709 |
+
"loss": 0.6879,
|
| 710 |
+
"step": 4500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.6053675926548732,
|
| 714 |
+
"eval_loss": 0.6919859647750854,
|
| 715 |
+
"eval_runtime": 407.9173,
|
| 716 |
+
"eval_samples_per_second": 122.763,
|
| 717 |
+
"eval_steps_per_second": 1.92,
|
| 718 |
+
"step": 4500
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.6120938992399274,
|
| 722 |
+
"grad_norm": 1.310651421546936,
|
| 723 |
+
"learning_rate": 3.313920386142892e-06,
|
| 724 |
+
"loss": 0.6857,
|
| 725 |
+
"step": 4550
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.6188202058249815,
|
| 729 |
+
"grad_norm": 1.2892873287200928,
|
| 730 |
+
"learning_rate": 3.272542485937369e-06,
|
| 731 |
+
"loss": 0.6812,
|
| 732 |
+
"step": 4600
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 0.6255465124100357,
|
| 736 |
+
"grad_norm": 1.2226920127868652,
|
| 737 |
+
"learning_rate": 3.230929261806842e-06,
|
| 738 |
+
"loss": 0.6774,
|
| 739 |
+
"step": 4650
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 0.6322728189950898,
|
| 743 |
+
"grad_norm": 1.2619507312774658,
|
| 744 |
+
"learning_rate": 3.189093389542498e-06,
|
| 745 |
+
"loss": 0.6832,
|
| 746 |
+
"step": 4700
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.6389991255801439,
|
| 750 |
+
"grad_norm": 1.4155713319778442,
|
| 751 |
+
"learning_rate": 3.147047612756302e-06,
|
| 752 |
+
"loss": 0.6897,
|
| 753 |
+
"step": 4750
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.6457254321651981,
|
| 757 |
+
"grad_norm": 1.3170417547225952,
|
| 758 |
+
"learning_rate": 3.1048047389991693e-06,
|
| 759 |
+
"loss": 0.6962,
|
| 760 |
+
"step": 4800
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.6524517387502522,
|
| 764 |
+
"grad_norm": 1.3795112371444702,
|
| 765 |
+
"learning_rate": 3.062377635859663e-06,
|
| 766 |
+
"loss": 0.6838,
|
| 767 |
+
"step": 4850
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.6591780453353063,
|
| 771 |
+
"grad_norm": 1.3435876369476318,
|
| 772 |
+
"learning_rate": 3.019779227044398e-06,
|
| 773 |
+
"loss": 0.6813,
|
| 774 |
+
"step": 4900
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.6659043519203606,
|
| 778 |
+
"grad_norm": 1.4538263082504272,
|
| 779 |
+
"learning_rate": 2.9770224884413625e-06,
|
| 780 |
+
"loss": 0.6941,
|
| 781 |
+
"step": 4950
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.6726306585054147,
|
| 785 |
+
"grad_norm": 1.4418572187423706,
|
| 786 |
+
"learning_rate": 2.9341204441673267e-06,
|
| 787 |
+
"loss": 0.6901,
|
| 788 |
+
"step": 5000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 0.6726306585054147,
|
| 792 |
+
"eval_loss": 0.6878845691680908,
|
| 793 |
+
"eval_runtime": 407.8759,
|
| 794 |
+
"eval_samples_per_second": 122.775,
|
| 795 |
+
"eval_steps_per_second": 1.92,
|
| 796 |
+
"step": 5000
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 0.6793569650904688,
|
| 800 |
+
"grad_norm": 1.3608489036560059,
|
| 801 |
+
"learning_rate": 2.8910861626005774e-06,
|
| 802 |
+
"loss": 0.6822,
|
| 803 |
+
"step": 5050
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.686083271675523,
|
| 807 |
+
"grad_norm": 1.2994650602340698,
|
| 808 |
+
"learning_rate": 2.847932752400164e-06,
|
| 809 |
+
"loss": 0.6762,
|
| 810 |
+
"step": 5100
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 0.6928095782605771,
|
| 814 |
+
"grad_norm": 1.565314531326294,
|
| 815 |
+
"learning_rate": 2.804673358512869e-06,
|
| 816 |
+
"loss": 0.686,
|
| 817 |
+
"step": 5150
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.6995358848456312,
|
| 821 |
+
"grad_norm": 1.3825584650039673,
|
| 822 |
+
"learning_rate": 2.761321158169134e-06,
|
| 823 |
+
"loss": 0.6872,
|
| 824 |
+
"step": 5200
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.7062621914306854,
|
| 828 |
+
"grad_norm": 1.3237903118133545,
|
| 829 |
+
"learning_rate": 2.717889356869146e-06,
|
| 830 |
+
"loss": 0.6873,
|
| 831 |
+
"step": 5250
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 0.7129884980157396,
|
| 835 |
+
"grad_norm": 1.4130380153656006,
|
| 836 |
+
"learning_rate": 2.6743911843603134e-06,
|
| 837 |
+
"loss": 0.6849,
|
| 838 |
+
"step": 5300
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 0.7197148046007937,
|
| 842 |
+
"grad_norm": 1.3584030866622925,
|
| 843 |
+
"learning_rate": 2.6308398906073603e-06,
|
| 844 |
+
"loss": 0.6914,
|
| 845 |
+
"step": 5350
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 0.7264411111858479,
|
| 849 |
+
"grad_norm": 1.4434551000595093,
|
| 850 |
+
"learning_rate": 2.587248741756253e-06,
|
| 851 |
+
"loss": 0.6815,
|
| 852 |
+
"step": 5400
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 0.733167417770902,
|
| 856 |
+
"grad_norm": 1.343894362449646,
|
| 857 |
+
"learning_rate": 2.543631016093209e-06,
|
| 858 |
+
"loss": 0.6739,
|
| 859 |
+
"step": 5450
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 0.7398937243559561,
|
| 863 |
+
"grad_norm": 1.5112074613571167,
|
| 864 |
+
"learning_rate": 2.5e-06,
|
| 865 |
+
"loss": 0.6759,
|
| 866 |
+
"step": 5500
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.7398937243559561,
|
| 870 |
+
"eval_loss": 0.6843944787979126,
|
| 871 |
+
"eval_runtime": 407.3845,
|
| 872 |
+
"eval_samples_per_second": 122.923,
|
| 873 |
+
"eval_steps_per_second": 1.922,
|
| 874 |
+
"step": 5500
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 0.7466200309410103,
|
| 878 |
+
"grad_norm": 1.471646785736084,
|
| 879 |
+
"learning_rate": 2.4563689839067913e-06,
|
| 880 |
+
"loss": 0.6914,
|
| 881 |
+
"step": 5550
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.7533463375260644,
|
| 885 |
+
"grad_norm": 1.3042908906936646,
|
| 886 |
+
"learning_rate": 2.4127512582437486e-06,
|
| 887 |
+
"loss": 0.6813,
|
| 888 |
+
"step": 5600
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 0.7600726441111186,
|
| 892 |
+
"grad_norm": 1.2429543733596802,
|
| 893 |
+
"learning_rate": 2.3691601093926406e-06,
|
| 894 |
+
"loss": 0.6757,
|
| 895 |
+
"step": 5650
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 0.7667989506961728,
|
| 899 |
+
"grad_norm": 1.300581932067871,
|
| 900 |
+
"learning_rate": 2.325608815639687e-06,
|
| 901 |
+
"loss": 0.6861,
|
| 902 |
+
"step": 5700
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 0.7735252572812269,
|
| 906 |
+
"grad_norm": 1.3551619052886963,
|
| 907 |
+
"learning_rate": 2.2821106431308546e-06,
|
| 908 |
+
"loss": 0.677,
|
| 909 |
+
"step": 5750
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 0.780251563866281,
|
| 913 |
+
"grad_norm": 1.2355047464370728,
|
| 914 |
+
"learning_rate": 2.238678841830867e-06,
|
| 915 |
+
"loss": 0.6728,
|
| 916 |
+
"step": 5800
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 0.7869778704513352,
|
| 920 |
+
"grad_norm": 1.3812388181686401,
|
| 921 |
+
"learning_rate": 2.195326641487132e-06,
|
| 922 |
+
"loss": 0.6807,
|
| 923 |
+
"step": 5850
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 0.7937041770363893,
|
| 927 |
+
"grad_norm": 1.3730298280715942,
|
| 928 |
+
"learning_rate": 2.1520672475998374e-06,
|
| 929 |
+
"loss": 0.6835,
|
| 930 |
+
"step": 5900
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 0.8004304836214434,
|
| 934 |
+
"grad_norm": 1.3047642707824707,
|
| 935 |
+
"learning_rate": 2.1089138373994226e-06,
|
| 936 |
+
"loss": 0.695,
|
| 937 |
+
"step": 5950
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 0.8071567902064977,
|
| 941 |
+
"grad_norm": 1.2433747053146362,
|
| 942 |
+
"learning_rate": 2.0658795558326745e-06,
|
| 943 |
+
"loss": 0.6752,
|
| 944 |
+
"step": 6000
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 0.8071567902064977,
|
| 948 |
+
"eval_loss": 0.681221604347229,
|
| 949 |
+
"eval_runtime": 407.827,
|
| 950 |
+
"eval_samples_per_second": 122.79,
|
| 951 |
+
"eval_steps_per_second": 1.92,
|
| 952 |
+
"step": 6000
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 0.8138830967915518,
|
| 956 |
+
"grad_norm": 1.245276927947998,
|
| 957 |
+
"learning_rate": 2.022977511558638e-06,
|
| 958 |
+
"loss": 0.6822,
|
| 959 |
+
"step": 6050
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 0.8206094033766059,
|
| 963 |
+
"grad_norm": 1.306039810180664,
|
| 964 |
+
"learning_rate": 1.9802207729556023e-06,
|
| 965 |
+
"loss": 0.6902,
|
| 966 |
+
"step": 6100
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 0.8273357099616601,
|
| 970 |
+
"grad_norm": 1.3029533624649048,
|
| 971 |
+
"learning_rate": 1.937622364140338e-06,
|
| 972 |
+
"loss": 0.69,
|
| 973 |
+
"step": 6150
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 0.8340620165467142,
|
| 977 |
+
"grad_norm": 1.2798274755477905,
|
| 978 |
+
"learning_rate": 1.895195261000831e-06,
|
| 979 |
+
"loss": 0.6791,
|
| 980 |
+
"step": 6200
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 0.8407883231317683,
|
| 984 |
+
"grad_norm": 1.3335996866226196,
|
| 985 |
+
"learning_rate": 1.852952387243698e-06,
|
| 986 |
+
"loss": 0.6807,
|
| 987 |
+
"step": 6250
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 0.8475146297168225,
|
| 991 |
+
"grad_norm": 1.3937443494796753,
|
| 992 |
+
"learning_rate": 1.8109066104575023e-06,
|
| 993 |
+
"loss": 0.6817,
|
| 994 |
+
"step": 6300
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 0.8542409363018766,
|
| 998 |
+
"grad_norm": 1.3417977094650269,
|
| 999 |
+
"learning_rate": 1.7690707381931585e-06,
|
| 1000 |
+
"loss": 0.6824,
|
| 1001 |
+
"step": 6350
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 0.8609672428869308,
|
| 1005 |
+
"grad_norm": 1.2701561450958252,
|
| 1006 |
+
"learning_rate": 1.7274575140626318e-06,
|
| 1007 |
+
"loss": 0.674,
|
| 1008 |
+
"step": 6400
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 0.867693549471985,
|
| 1012 |
+
"grad_norm": 1.2082581520080566,
|
| 1013 |
+
"learning_rate": 1.686079613857109e-06,
|
| 1014 |
+
"loss": 0.6725,
|
| 1015 |
+
"step": 6450
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 0.8744198560570391,
|
| 1019 |
+
"grad_norm": 1.4414860010147095,
|
| 1020 |
+
"learning_rate": 1.6449496416858285e-06,
|
| 1021 |
+
"loss": 0.6826,
|
| 1022 |
+
"step": 6500
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 0.8744198560570391,
|
| 1026 |
+
"eval_loss": 0.6783373951911926,
|
| 1027 |
+
"eval_runtime": 407.3438,
|
| 1028 |
+
"eval_samples_per_second": 122.935,
|
| 1029 |
+
"eval_steps_per_second": 1.922,
|
| 1030 |
+
"step": 6500
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.8811461626420932,
|
| 1034 |
+
"grad_norm": 1.2324413061141968,
|
| 1035 |
+
"learning_rate": 1.6040801261367494e-06,
|
| 1036 |
+
"loss": 0.6682,
|
| 1037 |
+
"step": 6550
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 0.8878724692271474,
|
| 1041 |
+
"grad_norm": 1.2953166961669922,
|
| 1042 |
+
"learning_rate": 1.56348351646022e-06,
|
| 1043 |
+
"loss": 0.6811,
|
| 1044 |
+
"step": 6600
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.8945987758122015,
|
| 1048 |
+
"grad_norm": 1.2646501064300537,
|
| 1049 |
+
"learning_rate": 1.5231721787768162e-06,
|
| 1050 |
+
"loss": 0.6813,
|
| 1051 |
+
"step": 6650
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.9013250823972556,
|
| 1055 |
+
"grad_norm": 1.4454089403152466,
|
| 1056 |
+
"learning_rate": 1.4831583923105e-06,
|
| 1057 |
+
"loss": 0.6744,
|
| 1058 |
+
"step": 6700
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.9080513889823099,
|
| 1062 |
+
"grad_norm": 1.3868876695632935,
|
| 1063 |
+
"learning_rate": 1.443454345648252e-06,
|
| 1064 |
+
"loss": 0.6764,
|
| 1065 |
+
"step": 6750
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.914777695567364,
|
| 1069 |
+
"grad_norm": 1.2375882863998413,
|
| 1070 |
+
"learning_rate": 1.4040721330273063e-06,
|
| 1071 |
+
"loss": 0.678,
|
| 1072 |
+
"step": 6800
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 0.9215040021524181,
|
| 1076 |
+
"grad_norm": 1.3029659986495972,
|
| 1077 |
+
"learning_rate": 1.3650237506511333e-06,
|
| 1078 |
+
"loss": 0.6736,
|
| 1079 |
+
"step": 6850
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 0.9282303087374723,
|
| 1083 |
+
"grad_norm": 1.3237438201904297,
|
| 1084 |
+
"learning_rate": 1.3263210930352737e-06,
|
| 1085 |
+
"loss": 0.6769,
|
| 1086 |
+
"step": 6900
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 0.9349566153225264,
|
| 1090 |
+
"grad_norm": 1.3647958040237427,
|
| 1091 |
+
"learning_rate": 1.2879759493841577e-06,
|
| 1092 |
+
"loss": 0.6663,
|
| 1093 |
+
"step": 6950
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 0.9416829219075805,
|
| 1097 |
+
"grad_norm": 1.2964695692062378,
|
| 1098 |
+
"learning_rate": 1.2500000000000007e-06,
|
| 1099 |
+
"loss": 0.6804,
|
| 1100 |
+
"step": 7000
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 0.9416829219075805,
|
| 1104 |
+
"eval_loss": 0.6757956147193909,
|
| 1105 |
+
"eval_runtime": 407.4829,
|
| 1106 |
+
"eval_samples_per_second": 122.894,
|
| 1107 |
+
"eval_steps_per_second": 1.922,
|
| 1108 |
+
"step": 7000
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.9484092284926346,
|
| 1112 |
+
"grad_norm": 1.444977045059204,
|
| 1113 |
+
"learning_rate": 1.2124048127248644e-06,
|
| 1114 |
+
"loss": 0.6816,
|
| 1115 |
+
"step": 7050
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.9551355350776889,
|
| 1119 |
+
"grad_norm": 1.262220025062561,
|
| 1120 |
+
"learning_rate": 1.1752018394169882e-06,
|
| 1121 |
+
"loss": 0.6687,
|
| 1122 |
+
"step": 7100
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.961861841662743,
|
| 1126 |
+
"grad_norm": 1.2450040578842163,
|
| 1127 |
+
"learning_rate": 1.1384024124624324e-06,
|
| 1128 |
+
"loss": 0.6738,
|
| 1129 |
+
"step": 7150
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.9685881482477972,
|
| 1133 |
+
"grad_norm": 1.31246817111969,
|
| 1134 |
+
"learning_rate": 1.1020177413231334e-06,
|
| 1135 |
+
"loss": 0.6664,
|
| 1136 |
+
"step": 7200
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.9753144548328513,
|
| 1140 |
+
"grad_norm": 1.307934284210205,
|
| 1141 |
+
"learning_rate": 1.0660589091223854e-06,
|
| 1142 |
+
"loss": 0.6845,
|
| 1143 |
+
"step": 7250
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.9820407614179054,
|
| 1147 |
+
"grad_norm": 1.2466742992401123,
|
| 1148 |
+
"learning_rate": 1.0305368692688175e-06,
|
| 1149 |
+
"loss": 0.6732,
|
| 1150 |
+
"step": 7300
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.9887670680029595,
|
| 1154 |
+
"grad_norm": 1.3563283681869507,
|
| 1155 |
+
"learning_rate": 9.95462442119879e-07,
|
| 1156 |
+
"loss": 0.6683,
|
| 1157 |
+
"step": 7350
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.9954933745880137,
|
| 1161 |
+
"grad_norm": 1.344889760017395,
|
| 1162 |
+
"learning_rate": 9.608463116858544e-07,
|
| 1163 |
+
"loss": 0.6794,
|
| 1164 |
+
"step": 7400
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 1.002219681173068,
|
| 1168 |
+
"grad_norm": 1.2191548347473145,
|
| 1169 |
+
"learning_rate": 9.266990223754069e-07,
|
| 1170 |
+
"loss": 0.6551,
|
| 1171 |
+
"step": 7450
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 1.008945987758122,
|
| 1175 |
+
"grad_norm": 1.4573768377304077,
|
| 1176 |
+
"learning_rate": 8.930309757836517e-07,
|
| 1177 |
+
"loss": 0.6131,
|
| 1178 |
+
"step": 7500
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 1.008945987758122,
|
| 1182 |
+
"eval_loss": 0.6763768196105957,
|
| 1183 |
+
"eval_runtime": 407.3062,
|
| 1184 |
+
"eval_samples_per_second": 122.947,
|
| 1185 |
+
"eval_steps_per_second": 1.922,
|
| 1186 |
+
"step": 7500
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.0156722943431762,
|
| 1190 |
+
"grad_norm": 1.4787367582321167,
|
| 1191 |
+
"learning_rate": 8.598524275237321e-07,
|
| 1192 |
+
"loss": 0.5996,
|
| 1193 |
+
"step": 7550
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.0223986009282302,
|
| 1197 |
+
"grad_norm": 1.2250146865844727,
|
| 1198 |
+
"learning_rate": 8.271734841028553e-07,
|
| 1199 |
+
"loss": 0.6111,
|
| 1200 |
+
"step": 7600
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.0291249075132844,
|
| 1204 |
+
"grad_norm": 1.2959485054016113,
|
| 1205 |
+
"learning_rate": 7.950040998437541e-07,
|
| 1206 |
+
"loss": 0.5994,
|
| 1207 |
+
"step": 7650
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.0358512140983387,
|
| 1211 |
+
"grad_norm": 1.3388257026672363,
|
| 1212 |
+
"learning_rate": 7.633540738525066e-07,
|
| 1213 |
+
"loss": 0.6092,
|
| 1214 |
+
"step": 7700
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.0425775206833927,
|
| 1218 |
+
"grad_norm": 1.3289120197296143,
|
| 1219 |
+
"learning_rate": 7.322330470336314e-07,
|
| 1220 |
+
"loss": 0.6154,
|
| 1221 |
+
"step": 7750
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.049303827268447,
|
| 1225 |
+
"grad_norm": 1.4141819477081299,
|
| 1226 |
+
"learning_rate": 7.016504991533727e-07,
|
| 1227 |
+
"loss": 0.6132,
|
| 1228 |
+
"step": 7800
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.056030133853501,
|
| 1232 |
+
"grad_norm": 1.368107557296753,
|
| 1233 |
+
"learning_rate": 6.716157459520739e-07,
|
| 1234 |
+
"loss": 0.6147,
|
| 1235 |
+
"step": 7850
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.0627564404385552,
|
| 1239 |
+
"grad_norm": 1.3877215385437012,
|
| 1240 |
+
"learning_rate": 6.421379363065142e-07,
|
| 1241 |
+
"loss": 0.6234,
|
| 1242 |
+
"step": 7900
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.0694827470236092,
|
| 1246 |
+
"grad_norm": 1.3689961433410645,
|
| 1247 |
+
"learning_rate": 6.1322604944307e-07,
|
| 1248 |
+
"loss": 0.6041,
|
| 1249 |
+
"step": 7950
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.0762090536086635,
|
| 1253 |
+
"grad_norm": 1.4440808296203613,
|
| 1254 |
+
"learning_rate": 5.848888922025553e-07,
|
| 1255 |
+
"loss": 0.6012,
|
| 1256 |
+
"step": 8000
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.0762090536086635,
|
| 1260 |
+
"eval_loss": 0.6758261322975159,
|
| 1261 |
+
"eval_runtime": 407.4199,
|
| 1262 |
+
"eval_samples_per_second": 122.912,
|
| 1263 |
+
"eval_steps_per_second": 1.922,
|
| 1264 |
+
"step": 8000
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 1.0829353601937177,
|
| 1268 |
+
"grad_norm": 1.3043277263641357,
|
| 1269 |
+
"learning_rate": 5.571350963575728e-07,
|
| 1270 |
+
"loss": 0.6173,
|
| 1271 |
+
"step": 8050
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 1.0896616667787717,
|
| 1275 |
+
"grad_norm": 1.3491568565368652,
|
| 1276 |
+
"learning_rate": 5.299731159831953e-07,
|
| 1277 |
+
"loss": 0.6101,
|
| 1278 |
+
"step": 8100
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 1.096387973363826,
|
| 1282 |
+
"grad_norm": 1.3337068557739258,
|
| 1283 |
+
"learning_rate": 5.034112248817685e-07,
|
| 1284 |
+
"loss": 0.6157,
|
| 1285 |
+
"step": 8150
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 1.10311427994888,
|
| 1289 |
+
"grad_norm": 1.4220951795578003,
|
| 1290 |
+
"learning_rate": 4.774575140626317e-07,
|
| 1291 |
+
"loss": 0.613,
|
| 1292 |
+
"step": 8200
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 1.1098405865339342,
|
| 1296 |
+
"grad_norm": 1.3012259006500244,
|
| 1297 |
+
"learning_rate": 4.5211988927752026e-07,
|
| 1298 |
+
"loss": 0.6025,
|
| 1299 |
+
"step": 8250
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 1.1165668931189883,
|
| 1303 |
+
"grad_norm": 1.3978270292282104,
|
| 1304 |
+
"learning_rate": 4.27406068612396e-07,
|
| 1305 |
+
"loss": 0.6086,
|
| 1306 |
+
"step": 8300
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 1.1232931997040425,
|
| 1310 |
+
"grad_norm": 1.381862998008728,
|
| 1311 |
+
"learning_rate": 4.033235801364402e-07,
|
| 1312 |
+
"loss": 0.6044,
|
| 1313 |
+
"step": 8350
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 1.1300195062890968,
|
| 1317 |
+
"grad_norm": 1.3832786083221436,
|
| 1318 |
+
"learning_rate": 3.798797596089351e-07,
|
| 1319 |
+
"loss": 0.6117,
|
| 1320 |
+
"step": 8400
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 1.1367458128741508,
|
| 1324 |
+
"grad_norm": 1.3328065872192383,
|
| 1325 |
+
"learning_rate": 3.5708174824471947e-07,
|
| 1326 |
+
"loss": 0.6021,
|
| 1327 |
+
"step": 8450
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 1.143472119459205,
|
| 1331 |
+
"grad_norm": 1.5573444366455078,
|
| 1332 |
+
"learning_rate": 3.3493649053890325e-07,
|
| 1333 |
+
"loss": 0.6136,
|
| 1334 |
+
"step": 8500
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 1.143472119459205,
|
| 1338 |
+
"eval_loss": 0.675076425075531,
|
| 1339 |
+
"eval_runtime": 407.3808,
|
| 1340 |
+
"eval_samples_per_second": 122.924,
|
| 1341 |
+
"eval_steps_per_second": 1.922,
|
| 1342 |
+
"step": 8500
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 1.150198426044259,
|
| 1346 |
+
"grad_norm": 1.3546710014343262,
|
| 1347 |
+
"learning_rate": 3.134507321515107e-07,
|
| 1348 |
+
"loss": 0.6107,
|
| 1349 |
+
"step": 8550
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 1.1569247326293133,
|
| 1353 |
+
"grad_norm": 1.2918901443481445,
|
| 1354 |
+
"learning_rate": 2.9263101785268253e-07,
|
| 1355 |
+
"loss": 0.6085,
|
| 1356 |
+
"step": 8600
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 1.1636510392143673,
|
| 1360 |
+
"grad_norm": 1.3548731803894043,
|
| 1361 |
+
"learning_rate": 2.7248368952908055e-07,
|
| 1362 |
+
"loss": 0.6033,
|
| 1363 |
+
"step": 8650
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 1.1703773457994215,
|
| 1367 |
+
"grad_norm": 1.4083459377288818,
|
| 1368 |
+
"learning_rate": 2.53014884252083e-07,
|
| 1369 |
+
"loss": 0.623,
|
| 1370 |
+
"step": 8700
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 1.1771036523844756,
|
| 1374 |
+
"grad_norm": 1.3091658353805542,
|
| 1375 |
+
"learning_rate": 2.3423053240837518e-07,
|
| 1376 |
+
"loss": 0.6088,
|
| 1377 |
+
"step": 8750
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 1.1838299589695298,
|
| 1381 |
+
"grad_norm": 1.394978404045105,
|
| 1382 |
+
"learning_rate": 2.1613635589349756e-07,
|
| 1383 |
+
"loss": 0.6069,
|
| 1384 |
+
"step": 8800
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 1.190556265554584,
|
| 1388 |
+
"grad_norm": 1.3812838792800903,
|
| 1389 |
+
"learning_rate": 1.9873786636889908e-07,
|
| 1390 |
+
"loss": 0.6087,
|
| 1391 |
+
"step": 8850
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 1.197282572139638,
|
| 1395 |
+
"grad_norm": 1.3317595720291138,
|
| 1396 |
+
"learning_rate": 1.8204036358303173e-07,
|
| 1397 |
+
"loss": 0.6059,
|
| 1398 |
+
"step": 8900
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 1.2040088787246923,
|
| 1402 |
+
"grad_norm": 1.2979373931884766,
|
| 1403 |
+
"learning_rate": 1.6604893375699594e-07,
|
| 1404 |
+
"loss": 0.5978,
|
| 1405 |
+
"step": 8950
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 1.2107351853097463,
|
| 1409 |
+
"grad_norm": 1.427206039428711,
|
| 1410 |
+
"learning_rate": 1.507684480352292e-07,
|
| 1411 |
+
"loss": 0.6127,
|
| 1412 |
+
"step": 9000
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 1.2107351853097463,
|
| 1416 |
+
"eval_loss": 0.6747262477874756,
|
| 1417 |
+
"eval_runtime": 407.1101,
|
| 1418 |
+
"eval_samples_per_second": 123.006,
|
| 1419 |
+
"eval_steps_per_second": 1.923,
|
| 1420 |
+
"step": 9000
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 1.2174614918948006,
|
| 1424 |
+
"grad_norm": 1.2737078666687012,
|
| 1425 |
+
"learning_rate": 1.362035610017079e-07,
|
| 1426 |
+
"loss": 0.6171,
|
| 1427 |
+
"step": 9050
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 1.2241877984798548,
|
| 1431 |
+
"grad_norm": 1.4532371759414673,
|
| 1432 |
+
"learning_rate": 1.223587092621162e-07,
|
| 1433 |
+
"loss": 0.6093,
|
| 1434 |
+
"step": 9100
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 1.2309141050649088,
|
| 1438 |
+
"grad_norm": 1.4686241149902344,
|
| 1439 |
+
"learning_rate": 1.0923811009241142e-07,
|
| 1440 |
+
"loss": 0.6073,
|
| 1441 |
+
"step": 9150
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 1.237640411649963,
|
| 1445 |
+
"grad_norm": 1.3653959035873413,
|
| 1446 |
+
"learning_rate": 9.684576015420277e-08,
|
| 1447 |
+
"loss": 0.6131,
|
| 1448 |
+
"step": 9200
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 1.244366718235017,
|
| 1452 |
+
"grad_norm": 1.3024178743362427,
|
| 1453 |
+
"learning_rate": 8.518543427732951e-08,
|
| 1454 |
+
"loss": 0.6162,
|
| 1455 |
+
"step": 9250
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 1.2510930248200713,
|
| 1459 |
+
"grad_norm": 1.3427560329437256,
|
| 1460 |
+
"learning_rate": 7.426068431000883e-08,
|
| 1461 |
+
"loss": 0.6033,
|
| 1462 |
+
"step": 9300
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 1.2578193314051254,
|
| 1466 |
+
"grad_norm": 1.4388611316680908,
|
| 1467 |
+
"learning_rate": 6.407483803691216e-08,
|
| 1468 |
+
"loss": 0.6104,
|
| 1469 |
+
"step": 9350
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 1.2645456379901796,
|
| 1473 |
+
"grad_norm": 1.610689401626587,
|
| 1474 |
+
"learning_rate": 5.463099816548578e-08,
|
| 1475 |
+
"loss": 0.612,
|
| 1476 |
+
"step": 9400
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 1.2712719445752336,
|
| 1480 |
+
"grad_norm": 1.4857087135314941,
|
| 1481 |
+
"learning_rate": 4.593204138084006e-08,
|
| 1482 |
+
"loss": 0.6069,
|
| 1483 |
+
"step": 9450
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 1.2779982511602879,
|
| 1487 |
+
"grad_norm": 1.3649048805236816,
|
| 1488 |
+
"learning_rate": 3.798061746947995e-08,
|
| 1489 |
+
"loss": 0.6076,
|
| 1490 |
+
"step": 9500
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 1.2779982511602879,
|
| 1494 |
+
"eval_loss": 0.6744668483734131,
|
| 1495 |
+
"eval_runtime": 407.221,
|
| 1496 |
+
"eval_samples_per_second": 122.973,
|
| 1497 |
+
"eval_steps_per_second": 1.923,
|
| 1498 |
+
"step": 9500
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 1.284724557745342,
|
| 1502 |
+
"grad_norm": 1.2920268774032593,
|
| 1503 |
+
"learning_rate": 3.077914851215585e-08,
|
| 1504 |
+
"loss": 0.6195,
|
| 1505 |
+
"step": 9550
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 1.2914508643303961,
|
| 1509 |
+
"grad_norm": 1.3456374406814575,
|
| 1510 |
+
"learning_rate": 2.4329828146074096e-08,
|
| 1511 |
+
"loss": 0.5986,
|
| 1512 |
+
"step": 9600
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 1.2981771709154504,
|
| 1516 |
+
"grad_norm": 1.340410828590393,
|
| 1517 |
+
"learning_rate": 1.8634620896695044e-08,
|
| 1518 |
+
"loss": 0.6146,
|
| 1519 |
+
"step": 9650
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 1.3049034775005044,
|
| 1523 |
+
"grad_norm": 1.4664082527160645,
|
| 1524 |
+
"learning_rate": 1.3695261579316776e-08,
|
| 1525 |
+
"loss": 0.6148,
|
| 1526 |
+
"step": 9700
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 1.3116297840855586,
|
| 1530 |
+
"grad_norm": 1.3726171255111694,
|
| 1531 |
+
"learning_rate": 9.513254770636138e-09,
|
| 1532 |
+
"loss": 0.6112,
|
| 1533 |
+
"step": 9750
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 1.3183560906706129,
|
| 1537 |
+
"grad_norm": 1.3762577772140503,
|
| 1538 |
+
"learning_rate": 6.089874350439507e-09,
|
| 1539 |
+
"loss": 0.6066,
|
| 1540 |
+
"step": 9800
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 1.325082397255667,
|
| 1544 |
+
"grad_norm": 1.3914607763290405,
|
| 1545 |
+
"learning_rate": 3.4261631135654174e-09,
|
| 1546 |
+
"loss": 0.6114,
|
| 1547 |
+
"step": 9850
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 1.3318087038407211,
|
| 1551 |
+
"grad_norm": 1.2674617767333984,
|
| 1552 |
+
"learning_rate": 1.5229324522605949e-09,
|
| 1553 |
+
"loss": 0.6165,
|
| 1554 |
+
"step": 9900
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 1.3385350104257752,
|
| 1558 |
+
"grad_norm": 1.2815463542938232,
|
| 1559 |
+
"learning_rate": 3.8076210902182607e-10,
|
| 1560 |
+
"loss": 0.6101,
|
| 1561 |
+
"step": 9950
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 1.3452613170108294,
|
| 1565 |
+
"grad_norm": 1.4005900621414185,
|
| 1566 |
+
"learning_rate": 0.0,
|
| 1567 |
+
"loss": 0.6033,
|
| 1568 |
+
"step": 10000
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 1.3452613170108294,
|
| 1572 |
+
"eval_loss": 0.6744288802146912,
|
| 1573 |
+
"eval_runtime": 407.1493,
|
| 1574 |
+
"eval_samples_per_second": 122.994,
|
| 1575 |
+
"eval_steps_per_second": 1.923,
|
| 1576 |
+
"step": 10000
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 1.3452613170108294,
|
| 1580 |
+
"step": 10000,
|
| 1581 |
+
"total_flos": 1119551134941184.0,
|
| 1582 |
+
"train_loss": 0.6844510423541069,
|
| 1583 |
+
"train_runtime": 79339.5992,
|
| 1584 |
+
"train_samples_per_second": 16.133,
|
| 1585 |
+
"train_steps_per_second": 0.126
|
| 1586 |
+
}
|
| 1587 |
+
],
|
| 1588 |
+
"logging_steps": 50,
|
| 1589 |
+
"max_steps": 10000,
|
| 1590 |
+
"num_input_tokens_seen": 0,
|
| 1591 |
+
"num_train_epochs": 2,
|
| 1592 |
+
"save_steps": 500,
|
| 1593 |
+
"stateful_callbacks": {
|
| 1594 |
+
"TrainerControl": {
|
| 1595 |
+
"args": {
|
| 1596 |
+
"should_epoch_stop": false,
|
| 1597 |
+
"should_evaluate": false,
|
| 1598 |
+
"should_log": false,
|
| 1599 |
+
"should_save": true,
|
| 1600 |
+
"should_training_stop": true
|
| 1601 |
+
},
|
| 1602 |
+
"attributes": {}
|
| 1603 |
+
}
|
| 1604 |
+
},
|
| 1605 |
+
"total_flos": 1119551134941184.0,
|
| 1606 |
+
"train_batch_size": 8,
|
| 1607 |
+
"trial_name": null,
|
| 1608 |
+
"trial_params": null
|
| 1609 |
+
}
|