---
library_name: peft
license: apache-2.0
base_model: Qwen/Qwen3.5-9B
tags:
- base_model:adapter:Qwen/Qwen3.5-9B
- llama-factory
- lora
- transformers
pipeline_tag: text-generation
model-index:
- name: qwen3.5-9b-muspsy-fixed
  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. -->

# qwen3.5-9b-muspsy-fixed

This model is a fine-tuned version of [Qwen/Qwen3.5-9B](https://huggingface.co/Qwen/Qwen3.5-9B) on the muspsy_task1, the muspsy_task2 and the muspsy_task3_fixed datasets.
It achieves the following results on the evaluation set:
- Loss: 1.2583

## 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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.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_steps: 0.1
- num_epochs: 3.0

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 1.3583        | 0.3916 | 500  | 1.3389          |
| 1.2919        | 0.7833 | 1000 | 1.3028          |
| 1.2094        | 1.1747 | 1500 | 1.2874          |
| 1.2644        | 1.5663 | 2000 | 1.2754          |
| 1.2199        | 1.9579 | 2500 | 1.2583          |
| 1.1110        | 2.3493 | 3000 | 1.2692          |
| 1.1196        | 2.7410 | 3500 | 1.2652          |
| 1.1142        | 3.0    | 3831 | 1.2654          |


### Framework versions

- PEFT 0.18.1
- Transformers 5.8.0
- Pytorch 2.8.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2