metadata
base_model: []
library_name: transformers
tags:
- mergekit
- merge
grpo-cp1k
This is a merge of pre-trained language models created using mergekit.
Merge Details
Merge Method
This model was merged using the SLERP merge method.
Models Merged
The following models were included in the merge:
- /llm-data/models/Qwen3-32B
- /llm-data/models/Qwen3-32B-GRPO/ckpt-1000
Configuration
The following YAML configuration was used to produce this model:
base_model: /llm-data/models/Qwen3-32B
dtype: bfloat16
merge_method: slerp
modules:
default:
slices:
- sources:
- layer_range: [0, 64]
model: /llm-data/models/Qwen3-32B
- layer_range: [0, 64]
model: /llm-data/models/Qwen3-32B-GRPO/ckpt-1000
parameters:
t:
- filter: self_attn
value: [0.0, 0.5, 0.3, 0.7, 1.0]
- filter: mlp
value: [1.0, 0.5, 0.7, 0.3, 0.0]
- value: 0.5