---
base_model: Qwen/Qwen2-VL-2B-Instruct
library_name: peft
license: apache-2.0
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: qwen2-2b-instruct-trl-sft-mrg
  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. -->

# qwen2-2b-instruct-trl-sft-mrg

This model is a fine-tuned version of [Qwen/Qwen2-VL-2B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-2B-Instruct) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 1.2568

## 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: 0.0001
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 15

### Training results

| Training Loss | Epoch   | Step | Validation Loss |
|:-------------:|:-------:|:----:|:---------------:|
| 3.0236        | 0.4942  | 32   | 2.9521          |
| 2.6642        | 0.9884  | 64   | 2.4341          |
| 2.156         | 1.4846  | 96   | 1.9677          |
| 1.9011        | 1.9788  | 128  | 1.7312          |
| 1.6955        | 2.4749  | 160  | 1.6093          |
| 1.5552        | 2.9691  | 192  | 1.5437          |
| 1.5361        | 3.4653  | 224  | 1.4991          |
| 1.4831        | 3.9595  | 256  | 1.4554          |
| 1.5036        | 4.4556  | 288  | 1.4261          |
| 1.3815        | 4.9498  | 320  | 1.3991          |
| 1.3762        | 5.4459  | 352  | 1.3760          |
| 1.3636        | 5.9402  | 384  | 1.3562          |
| 1.2826        | 6.4363  | 416  | 1.3424          |
| 1.3178        | 6.9305  | 448  | 1.3256          |
| 1.2689        | 7.4266  | 480  | 1.3123          |
| 1.2163        | 7.9208  | 512  | 1.3019          |
| 1.284         | 8.4170  | 544  | 1.2920          |
| 1.3356        | 8.9112  | 576  | 1.2862          |
| 1.2359        | 9.4073  | 608  | 1.2820          |
| 1.2157        | 9.9015  | 640  | 1.2746          |
| 1.1936        | 10.3977 | 672  | 1.2709          |
| 1.3181        | 10.8919 | 704  | 1.2659          |
| 1.2266        | 11.3880 | 736  | 1.2641          |
| 1.213         | 11.8822 | 768  | 1.2605          |
| 1.1997        | 12.3784 | 800  | 1.2603          |
| 1.2584        | 12.8726 | 832  | 1.2577          |
| 1.2547        | 13.3687 | 864  | 1.2576          |
| 1.2544        | 13.8629 | 896  | 1.2574          |
| 1.203         | 14.3591 | 928  | 1.2569          |
| 1.1467        | 14.8533 | 960  | 1.2568          |


### Framework versions

- PEFT 0.11.1
- Transformers 4.46.3
- Pytorch 2.3.1+cu121
- Datasets 3.0.1
- Tokenizers 0.20.0