https://huggingface.co/NTQAI/Nxcode-CQ-7B-orpo
1,266 Pulls Updated 6 months ago
Updated 6 months ago
6 months ago
2784da3b3724 · 6.4GB
model
archqwen2
·
parameters7.25B
·
quantizationQ6_K
6.4GB
params
{"stop":["\u003c|im_start|\u003e","\u003c|im_end|\u003e","\u003c/s\u003e"]}
76B
template
{{ if .System }}<|im_start|>system
{{ .System }}<|im_end|>
{{ end }}{{ if .Prompt }}<|im_start|>us
186B
Readme
Introduction
Nxcode-CQ-7B-orpo is an Monolithic Preference Optimization without Reference Model fine-tune of Qwen/CodeQwen1.5-7B on 100k samples of high-quality ranking data.
Evalplus
| EvalPlus | pass@1 |
|---|---|
| HumanEval | 86.6 |
| HumanEval+ | 83.5 |
| MBPP(v0.2.0) | 82.3 |
| MBPP+(v0.2.0) | 70.4 |