tohuy2710
/

Qwen2.5-Coder-3B-Finetune-LeetcodeDataset

Model card Files Files and versions

Qwen/Qwen2.5-Coder-3B Full Finetune 5 epochs on newfacade/LeetCodeDataset
HumanEval Pass@1 improved from 0,52 to 0,62
Evaluation performance on benchmarks will be updated asap

Downloads last month: 4

Safetensors

Model size

3B params

Tensor type

BF16

·

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for tohuy2710/Qwen2.5-Coder-3B-Finetune-LeetcodeDataset

Base model

Qwen/Qwen2.5-3B

Finetuned

Qwen/Qwen2.5-Coder-3B

Finetuned

(43)

this model

Dataset used to train tohuy2710/Qwen2.5-Coder-3B-Finetune-LeetcodeDataset

Collection including tohuy2710/Qwen2.5-Coder-3B-Finetune-LeetcodeDataset

Qwen2.5-Coder Finetune LeetCodeDataset

Using LoRA, Full Finetune technique to finetune models in Qwen2.5-Coder series, improved score in several benchmarks (HumanEval, MBPP, DS1000, ...). • 4 items • Updated 11 days ago