Files
Qwen3-4B-Base-Continued-GRP…/mergekit_config.yml
ModelHub XC 913020eb1c 初始化项目,由ModelHub XC社区提供模型
Model: Lambent/Qwen3-4B-Base-Continued-GRPO-Merge
Source: Original Platform
2026-06-02 03:31:25 +08:00

10 lines
242 B
YAML

# TIES merge: Judge as base, inject sparse GRPO knowledge
merge_method: ties
base_model: ./merged_models/llm-judge-merged-fixed
models:
- model: ./merged_models/grpo-cabs
parameters:
density: 0.5
weight: 0.4
dtype: bfloat16