Model save
Browse files- README.md +68 -0
- all_results.json +9 -0
- generation_config.json +10 -0
- train_results.json +9 -0
- trainer_state.json +603 -0
README.md
ADDED
|
@@ -0,0 +1,68 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: llama3
|
| 4 |
+
base_model: meta-llama/Meta-Llama-3-8B
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
datasets:
|
| 10 |
+
- generator
|
| 11 |
+
model-index:
|
| 12 |
+
- name: llama3.1-8b-gpt4o_100k_summarize-fft
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# llama3.1-8b-gpt4o_100k_summarize-fft
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 3.8606
|
| 24 |
+
|
| 25 |
+
## Model description
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Intended uses & limitations
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training and evaluation data
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training procedure
|
| 38 |
+
|
| 39 |
+
### Training hyperparameters
|
| 40 |
+
|
| 41 |
+
The following hyperparameters were used during training:
|
| 42 |
+
- learning_rate: 0.0003
|
| 43 |
+
- train_batch_size: 2
|
| 44 |
+
- eval_batch_size: 2
|
| 45 |
+
- seed: 42
|
| 46 |
+
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 8
|
| 48 |
+
- gradient_accumulation_steps: 2
|
| 49 |
+
- total_train_batch_size: 32
|
| 50 |
+
- total_eval_batch_size: 16
|
| 51 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
+
- lr_scheduler_type: cosine
|
| 53 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 54 |
+
- num_epochs: 1
|
| 55 |
+
|
| 56 |
+
### Training results
|
| 57 |
+
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 60 |
+
| 0.9063 | 1.0 | 390 | 3.8606 |
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
### Framework versions
|
| 64 |
+
|
| 65 |
+
- Transformers 4.45.1
|
| 66 |
+
- Pytorch 2.4.1+cu121
|
| 67 |
+
- Datasets 3.0.1
|
| 68 |
+
- Tokenizers 0.20.0
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 20388343971840.0,
|
| 4 |
+
"train_loss": 1.4792061909651144,
|
| 5 |
+
"train_runtime": 8464.1713,
|
| 6 |
+
"train_samples": 115376,
|
| 7 |
+
"train_samples_per_second": 1.474,
|
| 8 |
+
"train_steps_per_second": 0.046
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 128256,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": 128257,
|
| 5 |
+
"max_length": 4096,
|
| 6 |
+
"pad_token_id": 128257,
|
| 7 |
+
"temperature": 0.6,
|
| 8 |
+
"top_p": 0.9,
|
| 9 |
+
"transformers_version": "4.45.1"
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 20388343971840.0,
|
| 4 |
+
"train_loss": 1.4792061909651144,
|
| 5 |
+
"train_runtime": 8464.1713,
|
| 6 |
+
"train_samples": 115376,
|
| 7 |
+
"train_samples_per_second": 1.474,
|
| 8 |
+
"train_steps_per_second": 0.046
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,603 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 390,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.002564102564102564,
|
| 13 |
+
"grad_norm": 32.246810828841255,
|
| 14 |
+
"learning_rate": 7.692307692307692e-06,
|
| 15 |
+
"loss": 2.2691,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.01282051282051282,
|
| 20 |
+
"grad_norm": 29.537508150180734,
|
| 21 |
+
"learning_rate": 3.8461538461538456e-05,
|
| 22 |
+
"loss": 2.5015,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.02564102564102564,
|
| 27 |
+
"grad_norm": 7.166984360831344,
|
| 28 |
+
"learning_rate": 7.692307692307691e-05,
|
| 29 |
+
"loss": 1.3116,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.038461538461538464,
|
| 34 |
+
"grad_norm": 7.358331315050627,
|
| 35 |
+
"learning_rate": 0.00011538461538461538,
|
| 36 |
+
"loss": 1.2847,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.05128205128205128,
|
| 41 |
+
"grad_norm": 8.061303504403417,
|
| 42 |
+
"learning_rate": 0.00015384615384615382,
|
| 43 |
+
"loss": 1.424,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0641025641025641,
|
| 48 |
+
"grad_norm": 30.923296803868965,
|
| 49 |
+
"learning_rate": 0.0001923076923076923,
|
| 50 |
+
"loss": 1.6455,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.07692307692307693,
|
| 55 |
+
"grad_norm": 35.007464110767536,
|
| 56 |
+
"learning_rate": 0.00023076923076923076,
|
| 57 |
+
"loss": 1.8041,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.08974358974358974,
|
| 62 |
+
"grad_norm": 10.894826428758686,
|
| 63 |
+
"learning_rate": 0.0002692307692307692,
|
| 64 |
+
"loss": 1.6061,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.10256410256410256,
|
| 69 |
+
"grad_norm": 32.31558367670942,
|
| 70 |
+
"learning_rate": 0.0002999939918069778,
|
| 71 |
+
"loss": 1.7147,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.11538461538461539,
|
| 76 |
+
"grad_norm": 408.39746177245365,
|
| 77 |
+
"learning_rate": 0.0002997837555846212,
|
| 78 |
+
"loss": 3.0391,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.1282051282051282,
|
| 83 |
+
"grad_norm": 585.894165849368,
|
| 84 |
+
"learning_rate": 0.00029927359084964874,
|
| 85 |
+
"loss": 7.5436,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.14102564102564102,
|
| 90 |
+
"grad_norm": 188.43190635567956,
|
| 91 |
+
"learning_rate": 0.00029846451916110426,
|
| 92 |
+
"loss": 6.3459,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.15384615384615385,
|
| 97 |
+
"grad_norm": 18.84113658139999,
|
| 98 |
+
"learning_rate": 0.00029735816061234965,
|
| 99 |
+
"loss": 4.155,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.16666666666666666,
|
| 104 |
+
"grad_norm": 83.76190854828539,
|
| 105 |
+
"learning_rate": 0.00029595673058697357,
|
| 106 |
+
"loss": 2.5163,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.1794871794871795,
|
| 111 |
+
"grad_norm": 8.013185246066367,
|
| 112 |
+
"learning_rate": 0.00029426303532268435,
|
| 113 |
+
"loss": 1.8248,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.19230769230769232,
|
| 118 |
+
"grad_norm": 7.809904167784529,
|
| 119 |
+
"learning_rate": 0.00029228046629207175,
|
| 120 |
+
"loss": 1.6297,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.20512820512820512,
|
| 125 |
+
"grad_norm": 6.240741545472446,
|
| 126 |
+
"learning_rate": 0.00029001299341148754,
|
| 127 |
+
"loss": 1.4935,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.21794871794871795,
|
| 132 |
+
"grad_norm": 56.36582267694748,
|
| 133 |
+
"learning_rate": 0.0002874651570916444,
|
| 134 |
+
"loss": 1.5049,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.23076923076923078,
|
| 139 |
+
"grad_norm": 244.98938960556112,
|
| 140 |
+
"learning_rate": 0.0002846420591458521,
|
| 141 |
+
"loss": 2.0434,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.24358974358974358,
|
| 146 |
+
"grad_norm": 15.923002750416765,
|
| 147 |
+
"learning_rate": 0.00028154935257409397,
|
| 148 |
+
"loss": 1.9592,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.2564102564102564,
|
| 153 |
+
"grad_norm": 6.887371068354224,
|
| 154 |
+
"learning_rate": 0.000278193230243403,
|
| 155 |
+
"loss": 1.52,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.2692307692307692,
|
| 160 |
+
"grad_norm": 2.1953075310025407,
|
| 161 |
+
"learning_rate": 0.00027458041248720175,
|
| 162 |
+
"loss": 1.4671,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.28205128205128205,
|
| 167 |
+
"grad_norm": 5.381841976743356,
|
| 168 |
+
"learning_rate": 0.0002707181336484383,
|
| 169 |
+
"loss": 1.7707,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.2948717948717949,
|
| 174 |
+
"grad_norm": 18.048940491315566,
|
| 175 |
+
"learning_rate": 0.00026661412759346485,
|
| 176 |
+
"loss": 1.4121,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.3076923076923077,
|
| 181 |
+
"grad_norm": 3.1798969853488384,
|
| 182 |
+
"learning_rate": 0.00026227661222566516,
|
| 183 |
+
"loss": 1.4291,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.32051282051282054,
|
| 188 |
+
"grad_norm": 43.65623626841003,
|
| 189 |
+
"learning_rate": 0.00025771427302984107,
|
| 190 |
+
"loss": 1.3341,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.3333333333333333,
|
| 195 |
+
"grad_norm": 26.333557314371472,
|
| 196 |
+
"learning_rate": 0.00025293624568031,
|
| 197 |
+
"loss": 1.7161,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.34615384615384615,
|
| 202 |
+
"grad_norm": 2.1036876520568257,
|
| 203 |
+
"learning_rate": 0.0002479520977475377,
|
| 204 |
+
"loss": 1.4303,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.358974358974359,
|
| 209 |
+
"grad_norm": 2.8179883662231773,
|
| 210 |
+
"learning_rate": 0.0002427718095399382,
|
| 211 |
+
"loss": 1.2882,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.3717948717948718,
|
| 216 |
+
"grad_norm": 2.0944018050914863,
|
| 217 |
+
"learning_rate": 0.00023740575411920265,
|
| 218 |
+
"loss": 1.3758,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.38461538461538464,
|
| 223 |
+
"grad_norm": 32.12093678906791,
|
| 224 |
+
"learning_rate": 0.00023186467652917567,
|
| 225 |
+
"loss": 1.3851,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.3974358974358974,
|
| 230 |
+
"grad_norm": 6.163138554388992,
|
| 231 |
+
"learning_rate": 0.00022615967227987015,
|
| 232 |
+
"loss": 1.3702,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.41025641025641024,
|
| 237 |
+
"grad_norm": 2.523613373490038,
|
| 238 |
+
"learning_rate": 0.00022030216512970553,
|
| 239 |
+
"loss": 1.278,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.4230769230769231,
|
| 244 |
+
"grad_norm": 4.836492744474631,
|
| 245 |
+
"learning_rate": 0.0002143038842104581,
|
| 246 |
+
"loss": 1.6826,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.4358974358974359,
|
| 251 |
+
"grad_norm": 35.30106381963297,
|
| 252 |
+
"learning_rate": 0.00020817684054072824,
|
| 253 |
+
"loss": 1.6584,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.44871794871794873,
|
| 258 |
+
"grad_norm": 5.821901622579396,
|
| 259 |
+
"learning_rate": 0.0002019333029749549,
|
| 260 |
+
"loss": 1.4078,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.46153846153846156,
|
| 265 |
+
"grad_norm": 1.8515370008819587,
|
| 266 |
+
"learning_rate": 0.00019558577363613702,
|
| 267 |
+
"loss": 1.1904,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.47435897435897434,
|
| 272 |
+
"grad_norm": 1.773287149962229,
|
| 273 |
+
"learning_rate": 0.00018914696288145557,
|
| 274 |
+
"loss": 1.1222,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.48717948717948717,
|
| 279 |
+
"grad_norm": 9.981307604755651,
|
| 280 |
+
"learning_rate": 0.0001826297638509251,
|
| 281 |
+
"loss": 1.1487,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.5,
|
| 286 |
+
"grad_norm": 3.292661747708301,
|
| 287 |
+
"learning_rate": 0.00017604722665003956,
|
| 288 |
+
"loss": 1.2224,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.5128205128205128,
|
| 293 |
+
"grad_norm": 1.5110397540810954,
|
| 294 |
+
"learning_rate": 0.00016941253221810829,
|
| 295 |
+
"loss": 1.164,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.5256410256410257,
|
| 300 |
+
"grad_norm": 1.809157946038207,
|
| 301 |
+
"learning_rate": 0.0001627389659346097,
|
| 302 |
+
"loss": 1.1763,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.5384615384615384,
|
| 307 |
+
"grad_norm": 1.3191427643036746,
|
| 308 |
+
"learning_rate": 0.00015603989101641228,
|
| 309 |
+
"loss": 1.2132,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.5512820512820513,
|
| 314 |
+
"grad_norm": 1.0948738027576763,
|
| 315 |
+
"learning_rate": 0.00014932872175913348,
|
| 316 |
+
"loss": 1.1136,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.5641025641025641,
|
| 321 |
+
"grad_norm": 39.38545507943537,
|
| 322 |
+
"learning_rate": 0.00014261889667621826,
|
| 323 |
+
"loss": 1.1053,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.5769230769230769,
|
| 328 |
+
"grad_norm": 6.565265116408364,
|
| 329 |
+
"learning_rate": 0.0001359238515895231,
|
| 330 |
+
"loss": 1.5669,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.5897435897435898,
|
| 335 |
+
"grad_norm": 4.60069460000724,
|
| 336 |
+
"learning_rate": 0.00012925699272529006,
|
| 337 |
+
"loss": 1.3127,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.6025641025641025,
|
| 342 |
+
"grad_norm": 2.5322952556318543,
|
| 343 |
+
"learning_rate": 0.0001226316698693831,
|
| 344 |
+
"loss": 1.189,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.6153846153846154,
|
| 349 |
+
"grad_norm": 1.4722453502945256,
|
| 350 |
+
"learning_rate": 0.0001160611496355417,
|
| 351 |
+
"loss": 1.1154,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.6282051282051282,
|
| 356 |
+
"grad_norm": 0.9179752397575098,
|
| 357 |
+
"learning_rate": 0.00010955858890017846,
|
| 358 |
+
"loss": 1.0788,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.6410256410256411,
|
| 363 |
+
"grad_norm": 1.6490754560884426,
|
| 364 |
+
"learning_rate": 0.00010313700845691636,
|
| 365 |
+
"loss": 1.5551,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.6538461538461539,
|
| 370 |
+
"grad_norm": 0.9038214719254013,
|
| 371 |
+
"learning_rate": 9.680926694361964e-05,
|
| 372 |
+
"loss": 1.0483,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.6666666666666666,
|
| 377 |
+
"grad_norm": 0.7506946296792257,
|
| 378 |
+
"learning_rate": 9.058803509412646e-05,
|
| 379 |
+
"loss": 1.0173,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.6794871794871795,
|
| 384 |
+
"grad_norm": 1.105603168554401,
|
| 385 |
+
"learning_rate": 8.448577036624309e-05,
|
| 386 |
+
"loss": 1.0209,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.6923076923076923,
|
| 391 |
+
"grad_norm": 0.7904976226318772,
|
| 392 |
+
"learning_rate": 7.851469199680381e-05,
|
| 393 |
+
"loss": 1.0113,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.7051282051282052,
|
| 398 |
+
"grad_norm": 0.8256317535886027,
|
| 399 |
+
"learning_rate": 7.268675653374722e-05,
|
| 400 |
+
"loss": 1.0168,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.717948717948718,
|
| 405 |
+
"grad_norm": 0.9927491874944325,
|
| 406 |
+
"learning_rate": 6.701363389420296e-05,
|
| 407 |
+
"loss": 0.999,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.7307692307692307,
|
| 412 |
+
"grad_norm": 0.73183193491982,
|
| 413 |
+
"learning_rate": 6.15066839965316e-05,
|
| 414 |
+
"loss": 0.9793,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.7435897435897436,
|
| 419 |
+
"grad_norm": 1.1245402624787457,
|
| 420 |
+
"learning_rate": 5.6176934013108364e-05,
|
| 421 |
+
"loss": 0.9756,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.7564102564102564,
|
| 426 |
+
"grad_norm": 0.6697033469064312,
|
| 427 |
+
"learning_rate": 5.103505628940178e-05,
|
| 428 |
+
"loss": 0.9658,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.7692307692307693,
|
| 433 |
+
"grad_norm": 0.7446212713719105,
|
| 434 |
+
"learning_rate": 4.609134697356009e-05,
|
| 435 |
+
"loss": 0.9734,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.782051282051282,
|
| 440 |
+
"grad_norm": 1.0552000330334352,
|
| 441 |
+
"learning_rate": 4.135570539930026e-05,
|
| 442 |
+
"loss": 0.9496,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.7948717948717948,
|
| 447 |
+
"grad_norm": 0.6168006421187829,
|
| 448 |
+
"learning_rate": 3.683761426338148e-05,
|
| 449 |
+
"loss": 0.9358,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.8076923076923077,
|
| 454 |
+
"grad_norm": 0.7790438783752558,
|
| 455 |
+
"learning_rate": 3.254612063735667e-05,
|
| 456 |
+
"loss": 1.0592,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.8205128205128205,
|
| 461 |
+
"grad_norm": 0.5966928897383734,
|
| 462 |
+
"learning_rate": 2.8489817851625024e-05,
|
| 463 |
+
"loss": 0.9039,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.8333333333333334,
|
| 468 |
+
"grad_norm": 11.644090734466477,
|
| 469 |
+
"learning_rate": 2.4676828288059558e-05,
|
| 470 |
+
"loss": 0.9481,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.8461538461538461,
|
| 475 |
+
"grad_norm": 0.676095723656071,
|
| 476 |
+
"learning_rate": 2.1114787115667476e-05,
|
| 477 |
+
"loss": 0.9089,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.8589743589743589,
|
| 482 |
+
"grad_norm": 0.7215500519599484,
|
| 483 |
+
"learning_rate": 1.7810827001850187e-05,
|
| 484 |
+
"loss": 0.9021,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.8717948717948718,
|
| 489 |
+
"grad_norm": 0.5589129407190109,
|
| 490 |
+
"learning_rate": 1.4771563829877597e-05,
|
| 491 |
+
"loss": 0.895,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.8846153846153846,
|
| 496 |
+
"grad_norm": 0.9293606746960693,
|
| 497 |
+
"learning_rate": 1.2003083451176366e-05,
|
| 498 |
+
"loss": 0.9188,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.8974358974358975,
|
| 503 |
+
"grad_norm": 0.5692953910303316,
|
| 504 |
+
"learning_rate": 9.510929498959269e-06,
|
| 505 |
+
"loss": 0.8814,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.9102564102564102,
|
| 510 |
+
"grad_norm": 0.6238210726565123,
|
| 511 |
+
"learning_rate": 7.300092287597742e-06,
|
| 512 |
+
"loss": 0.8627,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.9230769230769231,
|
| 517 |
+
"grad_norm": 0.5072463698927163,
|
| 518 |
+
"learning_rate": 5.374998819965653e-06,
|
| 519 |
+
"loss": 0.8626,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.9358974358974359,
|
| 524 |
+
"grad_norm": 0.6124784439233985,
|
| 525 |
+
"learning_rate": 3.7395039227639644e-06,
|
| 526 |
+
"loss": 0.8568,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.9487179487179487,
|
| 531 |
+
"grad_norm": 26.426978867070098,
|
| 532 |
+
"learning_rate": 2.3968825275764768e-06,
|
| 533 |
+
"loss": 0.8757,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.9615384615384616,
|
| 538 |
+
"grad_norm": 0.8103195703064432,
|
| 539 |
+
"learning_rate": 1.3498231131137293e-06,
|
| 540 |
+
"loss": 0.8811,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.9743589743589743,
|
| 545 |
+
"grad_norm": 0.557003321536524,
|
| 546 |
+
"learning_rate": 6.004223217757509e-07,
|
| 547 |
+
"loss": 0.8598,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.9871794871794872,
|
| 552 |
+
"grad_norm": 0.7470834443971104,
|
| 553 |
+
"learning_rate": 1.50180761314167e-07,
|
| 554 |
+
"loss": 0.8623,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.0,
|
| 559 |
+
"grad_norm": 0.6929955320341405,
|
| 560 |
+
"learning_rate": 0.0,
|
| 561 |
+
"loss": 0.9063,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.0,
|
| 566 |
+
"eval_loss": 3.8606326580047607,
|
| 567 |
+
"eval_runtime": 2.3303,
|
| 568 |
+
"eval_samples_per_second": 4.291,
|
| 569 |
+
"eval_steps_per_second": 0.429,
|
| 570 |
+
"step": 390
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.0,
|
| 574 |
+
"step": 390,
|
| 575 |
+
"total_flos": 20388343971840.0,
|
| 576 |
+
"train_loss": 1.4792061909651144,
|
| 577 |
+
"train_runtime": 8464.1713,
|
| 578 |
+
"train_samples_per_second": 1.474,
|
| 579 |
+
"train_steps_per_second": 0.046
|
| 580 |
+
}
|
| 581 |
+
],
|
| 582 |
+
"logging_steps": 5,
|
| 583 |
+
"max_steps": 390,
|
| 584 |
+
"num_input_tokens_seen": 0,
|
| 585 |
+
"num_train_epochs": 1,
|
| 586 |
+
"save_steps": 100,
|
| 587 |
+
"stateful_callbacks": {
|
| 588 |
+
"TrainerControl": {
|
| 589 |
+
"args": {
|
| 590 |
+
"should_epoch_stop": false,
|
| 591 |
+
"should_evaluate": false,
|
| 592 |
+
"should_log": false,
|
| 593 |
+
"should_save": true,
|
| 594 |
+
"should_training_stop": true
|
| 595 |
+
},
|
| 596 |
+
"attributes": {}
|
| 597 |
+
}
|
| 598 |
+
},
|
| 599 |
+
"total_flos": 20388343971840.0,
|
| 600 |
+
"train_batch_size": 2,
|
| 601 |
+
"trial_name": null,
|
| 602 |
+
"trial_params": null
|
| 603 |
+
}
|