This is a merge of pre-trained language models created using
mergekit.
This model was merged using the
DARE TIES merge method using
Gryphe/Pantheon-RP-1.0-8b-Llama-3 as a base.
1models:
2 - model: Sao10K/L3-8B-Stheno-v3.2
3 parameters:
4 density: 0.5
5 weight: 1.0
6 - model: Gryphe/Pantheon-RP-1.0-8b-Llama-3
7 parameters:
8 density: 0.6
9 weight: 1.0
10 - model: cgato/L3-TheSpice-8b-v0.8.3
11 parameters:
12 density: 0.4
13 weight: 0.75
14merge_method: dare_ties
15tokenizer_source: base
16base_model: Gryphe/Pantheon-RP-1.0-8b-Llama-3
17parameters:
18 int8_mask: true
19dtype: bfloat16
20
