My Thalassic series includes my favorite Llama 3 models with Deepseek R1, using various methods I made 4 merges. Thalassic Delta, which I made with an SCE merge, was the best of the set. I actually got some advice from Steelskull about how I had my top k parameter way too high. So I decided to lower it for this merge. Now because I feel Delta was still a decent model with the top k on 1, I halved the top k parameter to 0.50 which is basically the max recommended setting.
This is a merge of pre-trained language models created using
mergekit.
This model was merged using the
SCE merge method using
SicariusSicariiStuff/Negative_LLAMA_70B as a base.
1models:
2 # Pivot model
3 - model: deepseek-ai/DeepSeek-R1-Distill-Llama-70B
4 # Target models
5 - model: Sao10K/70B-L3.3-Cirrus-x1
6 - model: Sao10K/L3.1-70B-Hanami-x1
7 - model: TheDrummer/Anubis-70B-v1
8 - model: EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1
9merge_method: sce
10base_model: SicariusSicariiStuff/Negative_LLAMA_70B
11parameters:
12 select_topk: 0.5
13dtype: bfloat16
