radna/Fused-Fuse-DeepSeek-R1-32B-RL-S1-20-52

Fused-Fuse-DeepSeek-R1-32B-RL-S1-20+52

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the sce merge method using FuseAI/FuseO1-DeepSeekR1-QwQ-SkyT1-32B-Preview as a base.

Models Merged

The following models were included in the merge:

• radna/Fuse-DeepSeek-R1-32B-RL-S1-20
• radna/Fuse-DeepSeek-R1-32B-RL-S1-52

Configuration

The following YAML configuration was used to produce this model:

models:
  # Pivot model
  - model: FuseAI/FuseO1-DeepSeekR1-QwQ-SkyT1-32B-Preview
  # Target models
  - model: radna/Fuse-DeepSeek-R1-32B-RL-S1-20
  - model: radna/Fuse-DeepSeek-R1-32B-RL-S1-52
merge_method: sce
base_model: FuseAI/FuseO1-DeepSeekR1-QwQ-SkyT1-32B-Preview
parameters:
  select_topk: 1.0
dtype: bfloat16