---
library_name: peft
license: apache-2.0
base_model: Qwen/Qwen2.5-7B
tags:
- axolotl
- generated_from_trainer
datasets:
- Aratako/Magpie-Tanuki-8B-annotated-96k
- Aratako/Synthetic-JP-EN-Coding-Dataset-Magpie-69k
- DataPilot/Zero_SFT_Ja_v2_b3t4
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
model-index:
- name: Qwen2.5-7B-axolotl-sft-v0.2
  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. -->

[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>

axolotl version: `0.8.0`
```yaml
base_model: Qwen/Qwen2.5-7B
hub_model_id: OsakanaTeishoku/Qwen2.5-7B-axolotl-sft-v0.2

load_in_8bit: false
load_in_4bit: true
strict: false

chat_template: qwen_25

datasets:
  # This will be the path used for the data when it is saved to the Volume in the cloud.
  - path: Aratako/Magpie-Tanuki-8B-annotated-96k
    split: train
    type: chat_template
    field_messages: messages
  - path: Aratako/Synthetic-JP-EN-Coding-Dataset-Magpie-69k
    split: train
    type: chat_template
    field_messages: messages
  - path: DataPilot/Zero_SFT_Ja_v2_b3t4
    split: train
    type: chat_template
    field_messages: conversation
    message_property_mappings:
      role: from
      content: value

shuffle_merged_datasets: true

dataset_prepared_path: last_run_prepared
#val_set_size: 0.05
output_dir: ./lora-out

sequence_len: 2048
sample_packing: false
eval_sample_packing: false
pad_to_sequence_len: false

adapter: qlora
lora_model_dir:
lora_r: 16
lora_alpha: 32
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:
lora_modules_to_save: # required when adding new tokens to LLaMA/Mistral
  - embed_tokens
  - lm_head

wandb_project: modal-axolotl
wandb_name: 20250419-qwen7b-modal

gradient_accumulation_steps: 4
micro_batch_size: 16
#auto_find_batch_size: true
#num_epochs: 1
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0001

bf16: true
fp16: false
tf32: false
train_on_inputs: false
group_by_length: false

gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention: true
flash_attention: 

warmup_ratio: 0.05
save_steps: 50
max_steps: 200
debug:
#deepspeed: /workspace/axolotl/deepspeed_configs/zero2.json
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
  eos_token: "<|im_end|>"

plugins:
  - axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: true
liger_glu_activation: true
liger_layer_norm: true
liger_fused_linear_cross_entropy: true

eval_strategy: "no"
save_strategy: "steps"
```

</details><br>

# Qwen2.5-7B-axolotl-sft-v0.2

This model is a fine-tuned version of [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B) on the Aratako/Magpie-Tanuki-8B-annotated-96k, the Aratako/Synthetic-JP-EN-Coding-Dataset-Magpie-69k and the DataPilot/Zero_SFT_Ja_v2_b3t4 datasets.

## 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: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Use OptimizerNames.ADAMW_BNB 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: 10
- training_steps: 200

### Training results



### Framework versions

- PEFT 0.15.1
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1