---
base_model: meta-llama/Llama-3.1-8B-Instruct
library_name: peft
license: llama3.1
tags:
- alignment-handbook
- trl
- sft
- generated_from_trainer
model-index:
- name: llama_3_gsm8k_midset_cot_simplest
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# llama_3_gsm8k_midset_cot_simplest

This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5414

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 200

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 1.121         | 0.04   | 5    | 1.0499          |
| 0.7969        | 0.08   | 10   | 0.7587          |
| 0.7177        | 0.12   | 15   | 0.6761          |
| 0.6315        | 0.16   | 20   | 0.6275          |
| 0.5984        | 0.2    | 25   | 0.5965          |
| 0.5678        | 0.24   | 30   | 0.5893          |
| 0.6067        | 0.28   | 35   | 0.5732          |
| 0.6306        | 0.32   | 40   | 0.5717          |
| 0.6134        | 0.36   | 45   | 0.5688          |
| 0.5396        | 0.4    | 50   | 0.5616          |
| 0.5796        | 0.44   | 55   | 0.5596          |
| 0.582         | 0.48   | 60   | 0.5543          |
| 0.5793        | 0.52   | 65   | 0.5518          |
| 0.4974        | 0.56   | 70   | 0.5556          |
| 0.5791        | 0.6    | 75   | 0.5544          |
| 0.5484        | 0.64   | 80   | 0.5518          |
| 0.5611        | 0.68   | 85   | 0.5500          |
| 0.5736        | 0.72   | 90   | 0.5474          |
| 0.5269        | 0.76   | 95   | 0.5490          |
| 0.58          | 0.8    | 100  | 0.5485          |
| 0.5295        | 0.84   | 105  | 0.5476          |
| 0.5124        | 0.88   | 110  | 0.5447          |
| 0.5368        | 0.92   | 115  | 0.5409          |
| 0.5705        | 0.96   | 120  | 0.5408          |
| 0.5244        | 1.0    | 125  | 0.5394          |
| 0.4334        | 1.04   | 130  | 0.5402          |
| 0.5256        | 1.08   | 135  | 0.5425          |
| 0.4398        | 1.12   | 140  | 0.5464          |
| 0.4843        | 1.16   | 145  | 0.5472          |
| 0.4437        | 1.2    | 150  | 0.5457          |
| 0.5133        | 1.24   | 155  | 0.5449          |
| 0.497         | 1.28   | 160  | 0.5427          |
| 0.4682        | 1.32   | 165  | 0.5420          |
| 0.4872        | 1.3600 | 170  | 0.5415          |
| 0.4763        | 1.4    | 175  | 0.5413          |
| 0.4647        | 1.44   | 180  | 0.5417          |
| 0.4679        | 1.48   | 185  | 0.5417          |
| 0.476         | 1.52   | 190  | 0.5414          |
| 0.462         | 1.56   | 195  | 0.5414          |
| 0.4665        | 1.6    | 200  | 0.5414          |


### Framework versions

- PEFT 0.12.0
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.0
- Tokenizers 0.19.1