This is a merge of pre-trained language models created using
mergekit.
This model was merged using the
DARE TIES merge method using
tinycompany/Shawty-1.4B-SFT-Stage-1 as a base.
1models:
2 - model: tinycompany/Shawty-1.4B-SFT-Stage-1
3 parameters:
4 density: 1.0
5 weight: 0.7
6 - model: tinycompany/shawty-CoT-Hindi-English
7 parameters:
8 density: 0.3
9 weight: 0.3
10merge_method: dare_ties
11base_model: tinycompany/Shawty-1.4B-SFT-Stage-1
12parameters:
13 normalize: true
14 int8_mask: true
15dtype: bfloat16
16tokenizer_source: base
17
