Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- Due to hardware limitation, some merge was done in 2 part.
- - Recreate (private) ReML: Mythologic (v2) (Chronos/Hermes/Airboros)
- => Replacing Chronos by The-Face-Of-Goonery/Chronos-Beluga-v2-13bfp16 (0.30)
- => Replacing Airoboros by jondurbin/airoboros-l2-13b-2.1 (last version) (0.40)
- => Keeping NousResearch/Nous-Hermes-Llama2-13b (0.30)
- Part 1: python ties_merge.py TheBloke/Llama-2-13B-fp16 ./ReML-L2-13B-part1 --merge The-Face-Of-Goonery/Chronos-Beluga-v2-13bfp16 --density 0.42 --merge jondurbin/airoboros-l2-13b-2.1 --density 0.56 --cuda
- Part 2: python ties_merge.py TheBloke/Llama-2-13B-fp16 ./ReML-L2-13B --merge NousResearch/Nous-Hermes-Llama2-13b --density 0.30 --merge Undi95/ReML-L2-13B-part1 --density 0.70 --cuda
- With that :
- - Recreate (private for now) ReMM: MythoMax (v2) (Mythologic/Huginn v1)
- => Replacing Mythologic by the one above (0.5)
- => Replacing Huginn by The-Face-Of-Goonery/Huginn-13b-v1.2 (hottest) (0.5)
- python ties_merge.py TheBloke/Llama-2-13B-fp16 ./ReMM-L2-13B --merge Undi95/Mythologic-L2-v2-13B --density 0.50 --merge The-Face-Of-Goonery/Huginn-13b-v1.2 --density 0.50 --cuda
- Last mix :
- - MythoMax (v2) (Base) (0.66)
- - Doctor-Shotgun/llama-2-13b-chat-limarp-v2-merged (Llama Chat Uncensored) (0.25)
- - KoboldAI/LLAMA2-13B-Holodeck-1 (0.09)
- Part 1: python ties_merge.py TheBloke/Llama-2-13B-fp16 ./MLewdBase-L2-13B-part1 --merge Undi95/ReMM-L2-13B --density 0.88 --merge KoboldAI/LLAMA2-13B-Holodeck-1 --density 0.12 --cuda
- Part 2: python ties_merge.py TheBloke/Llama-2-13B-fp16 ./MLewdBase-L2-13B --merge Undi95/MLewdBase-L2-13B-part1 --density 0.75 --merge Doctor-Shotgun/llama-2-13b-chat-limarp-v2-merged --density 0.25 --cuda
- Final touch :
- - Applying LoRA: nRuaif/Kimiko-v2-13B at (0.15) weight
- => Result: MLewd-L2-13B-v1-1
- ================== RETRY ===========================
- Last mix :
- - ReMM (Base) (0.57)
- - Doctor-Shotgun/llama-2-13b-chat-limarp-v2-merged (Llama Chat Uncensored) (0.35)
- - KoboldAI/LLAMA2-13B-Holodeck-1 (0.08)
- Part 1: python ties_merge.py TheBloke/Llama-2-13B-fp16 ./MLewdBase-L2-13B-part1 --merge Undi95/ReMM-L2-13B --density 0.88 --merge KoboldAI/LLAMA2-13B-Holodeck-1 --density 0.12 --cuda
- Part 2: python ties_merge.py TheBloke/Llama-2-13B-fp16 ./MLewdBase-L2-13B --merge Undi95/MLewdBase-L2-13B-part1 --density 0.65 --merge Doctor-Shotgun/llama-2-13b-chat-limarp-v2-merged --density 0.35 --cuda
- (MLewd-L2-13B-v1-2 got disqualified)
- - Applying LoRA: nRuaif/Kimiko-v2-13B at (0.24) weight on MLewd-L2-13B-v1-1
- => Result: MLewd-L2-13B-v1-3
- Remix:
- - Undi95/MLewd-L2-13B-v1-3 (0.81)
- - kajdun/iubaris-13b-v3 (0.9)
- - IkariDev/Athena-v1 (0.10)
- Part 1 : !python ties_merge.py TheBloke/Llama-2-13B-fp16 ./MLewd-L2-13B-v1-4-part1 --merge Undi95/MLewd-L2-13B-v1-3 --density 0.90 --merge kajdun/iubaris-13b-v3 --density 0.10 --cuda
- Part 2 : !python ties_merge.py TheBloke/Llama-2-13B-fp16 ./MLewd-L2-13B-v1-4 --merge Undi95/MLewd-L2-13B-v1-4-part1 --density 0.90 --merge IkariDev/Athena-v1 --density 0.10 --cuda
- => Result: MLewd-L2-13B-v1-4
- Hopeless attempt:
- - Undi95/MLewd-L2-13B-v1-4 (0.81)
- - Sao10K/Stheno-Inverted-L2-13B (0.10)
- (MLewd-L2-13B-v1-5 got disqualified)
- !python ties_merge.py TheBloke/Llama-2-13B-fp16 ./MLewd-L2-13B-v1-6 --merge Undi95/MLewd-L2-13B-v1-4 --density 0.90 --merge Sao10K/Stheno-Inverted-L2-13B --density 0.10 --cuda
- => Result: MLewd-L2-13B-v1-6
- ================== ERP RANKING ===========================
- 19.42 | MLewd-L2-13B-v1-3.q5_K_M.gguf (-> Best)
- 19.25 | MLewd-L2-13B-v1-1.q5_K_M.gguf
- 18.58 | MLewd-L2-13B-v1-5.q5_K_M.gguf
- 18.25 | MLewd-L2-13B-v1-2.q5 K M.gguf
- 17.92 | MLewd-L2-13B-v1-6.q5 K M.aauf
- 17.92 | MLewd-L2-13B-v1-4.q5_K_M.gguf
- ================== existential crisis ===========================
- What do I do with my life ? Do I really need Lewd ?
- Yes. Yes ? No. Yes. Maybe. Yes. Lewd. It's been 3 days.
- Help me. I'm horny.
- ================== RETRY 2 ===========================
- Mix:
- - Undi95/MLewd-L2-13B-v1-3 (0.82)
- - Sao10K/Stheno-Inverted-L2-13B (0.18)
- !python ties_merge.py TheBloke/Llama-2-13B-fp16 ./MLewd-L2-13B-v1-7 --merge Undi95/MLewd-L2-13B-v1-3 --density 0.82 --merge Sao10K/Stheno-Inverted-L2-13B --density 0.18 --cuda
- => Result: MLewd-L2-13B-v1-7
- Final touch (trying my best here) :
- Trial1: MLewd-L2-13B-v1-7 (0.77) + usernamedesu/MythKimikoBlue-Mix-v2-lora (LoRA 0.23) => MLewd-L2-13B-v1-7-TRY
- Trial2: MLewd-L2-13B-v1-7 (0.77) + zarakiquemparte/PIPPA-ShareGPT-Subset-QLora-13b (LoRA 0.23) => MLewd-L2-13B-v1-7-TRY2
- Endmysuffering v69: MLewd-L2-13B-v1-7-TRY (0.80) + zarakiquemparte/PIPPA-ShareGPT-Subset-QLora-13b (LoRA 0.20)
- => MLewd-L2-13B-v1-8
- Endmysuffering v96: MLewd-L2-13B-v1-7-TRY2 (0.80) + usernamedesu/MythKimikoBlue-Mix-v2-lora (LoRA 0.20)
- => MLewd-L2-13B-v1-8-2
- FINAL : MLewd-L2-13B-v1-7-TRY2 (0.82) + BluemoonRP (0.18)
- (LoRA 0.20) => MLewd-L2-13B-v1-8-3
- Undi95/MLewd-L2-13B-v1-3 (0.82) + zarakiquemparte/PIPPA-ShareGPT-Subset-QLora-13b (0.18) = MLewd-L2-13B-v1-3-PIPPA
- Undi95/ReMM-L2-13B (0.82) + zarakiquemparte/PIPPA-ShareGPT-Subset-QLora-13b (0.18) = ReMM-L2-13B-PIPPA
Add Comment
Please, Sign In to add comment