This is a merge of pre-trained language models created using
mergekit.
This model was merged using the
TIES merge method using
deepseek-ai/DeepSeek-R1-Distill-Llama-8B as a base.
1models:
2 - model: deepseek-ai/DeepSeek-R1-Distill-Llama-8B
3 parameters:
4 weight: 1
5 density: 1
6 - model: nvidia/Llama-3.1-Nemotron-8B-UltraLong-4M-Instruct
7 parameters:
8 weight: 1
9 density: 1
10merge_method: ties
11base_model: deepseek-ai/DeepSeek-R1-Distill-Llama-8B
12parameters:
13 density: 1
14 normalize: true
15 int8_mask: true
16dtype: bfloat16
17
