metadata
library_name: transformers
tags:
- small-lm
- math
- reasoning
- slm
license: apache-2.0
datasets:
- openai/gsm8k
base_model:
- Qwen/Qwen3-0.6B
Qwen3-0.6B-Math
This model is obtained by fine-tuning Qwen/Qwen3-0.6B on the gsm8k train split. The model is used in the experiments described in https://bknyaz.github.io/blog/2026/meta-merge/. Single A100 was used for fine-tuning and evaluation.
The following versions were used for train/eval:
- python >= 3.10
- torch : 2.9.0+cu128
- lm_eval : 0.4.9.1
- vllm : 0.11.1
- transformers : 4.57.6
- datasets : 3.2.0
- numpy : 2.2.6
Training
The TRL library was used with SFT/full-rank options:
python trl/scripts/sft.py --model_name_or_path Qwen/Qwen3-0.6B --dataset_name openai/gsm8k --dataset_config main --learning_rate 2e-5 \
--num_train_epochs 1 --per_device_train_batch_size 2 --gradient_accumulation_steps 8 --gradient_checkpointing --eos_token '<|im_end|>' --eval_strategy steps \
--eval_steps 100 --completion_only_loss True --report_to wandb --output_dir /path/to/the/finetuned/model
This is by far not the most compute and performance efficient fine-tuning, but it could be a good baseline.
The dataset was preprocessed to the conversational format:
# trl/scripts/sft.py
dataset = load_dataset(...)
def preprocess_function(example):
return {
"prompt": [{"role": "user", "content": example["question"]}],
"completion": [
{"role": "assistant", "content": example['answer']}
],
}
dataset = dataset.map(preprocess_function)
Evaluation
Evaluation was done with lm_eval on the test split of gsm8k:
python -m lm_eval --model vllm --model_args pretrained=${model},tensor_parallel_size=1,dtype=auto,gpu_memory_utilization=0.9,data_parallel_size=1 \
--tasks gsm8k --batch_size 1 --apply_chat_template=True --confirm_run_unsafe_code --trust_remote_code
Results
| Model | gsm8k |
|---|---|
| Qwen3-0.6B | 21.0 |
| Qwen3-0.6B-Math | 46.3 |
License
Please refer to the license of the original model Qwen/Qwen3-0.6B and dataset gsm8k.