metadata
license: other
license_name: deepseek
license_link: https://github.com/deepseek-ai/DeepSeek-Coder/blob/main/LICENSE-MODEL
(Note: From short testing, this Alt version generated much better code)
Alternate version of DeepMagic-Coder-7b which can be found bellow.
This version uses a diffrent config setup, with the actual base model of the two merges as the "base_model". Test both for yourself and see which is better at coding. Benchmarks coming soon.
Config can be found bellow:
models:
- model: deepseek-ai_deepseek-coder-6.7b-instruct
parameters:
weight: 1
- model: ise-uiuc_Magicoder-S-DS-6.7B
parameters:
weight: 1
merge_method: task_arithmetic
base_model: deepseek-ai_deepseek-coder-6.7b-base
parameters:
normalize: true
int8_mask: true
dtype: float16