[!CAUTION]
⚠️ Warning: This model can produce narratives and RP that contain violent and graphic erotic content. Adjust your system prompt accordingly. Also, use Mistral non-Tekken for best results.
StormSeeker 24B v1
ss7
This took 14 hours to merge using a custom method.
According to the audit, Loki, PaintedFantasy, and Hearthfire had a bit more influence than the other 3 models, but nothing was "drowned out".
The model is rather uncensored even without ablation and responds to some (not all) harmful prompt without refusals or jailbreaks, so a light jailbreak works effectively to bypass most censorship.
1models:2-model: A:\LLM\.cache\huggingface\hub\!models--CrucibleLab--M3.2-24B-Loki-V23-model: A:\LLM\.cache\huggingface\hub\!models--LatitudeGames--Hearthfire-24B4-model: A:\LLM\.cache\huggingface\hub\!models--PocketDoc--Dans-PersonalityEngine-V1.3.0-24b5-model: A:\LLM\.cache\huggingface\hub\!models--ReadyArt--Dark-Nexus-24B-v2.06-model: A:\LLM\.cache\huggingface\hub\!models--TheDrummer--Cydonia-24B-v4.37-model: A:\LLM\.cache\huggingface\hub\!models--zerofata--MS3.2-PaintedFantasy-v3-24B8merge_method: flux # version 5, Y6 config9parameters:10resume_path:"A:/mergekit-main/Storm_Cache"11tol:1e-912max_iter:1005#maximum BF16 fidelity13kappa:0.814eta:0.915auto_buffer:true16dtype: float32
17out_dtype: bfloat16
18tokenizer:19source: union
20chat_template: auto
21name: StormSeeker-24B-v1
According to the LLM, quantizations made from merges with merge_method: flux benefit greatly from the smaller block sizes of IQ4_NL, saying it should be on par with or slightly better than Q6_K, despite having higher perplexity. Note that this doesn't apply to merges made using standard methods like model_stock or karcher, and this claim has not yet been empirically verified.
Due to space/time constraints I am only uploading the following GGUFs:
