Roleplay Quantization in EXL2 format for Lumimaid 0.2 70B
Quantized using the cleaned PIPPA roleplay dataset.
2.0bpw8h quants (tested and barely working on one RTX 3090 at 32k context length)
3.75bpw8h quants (tested and working on dual RTX 3090s at 128k context length)
4.0bpw8h quants (tested and working on dual RTX 3090s at 98k context length)
4.4bpw8h quants (tested and working on dual RTX 3090s at 64k context length)
All tests performed on a headless Linux instance with no active desktop environment to maximize VRAM. Software used was TabbyAPI with Q4 cache enabled.
Other quants available on request, feel free to ask!
See original model for further details.
Original Model card
Lumimaid 0.2
This model is based on: Meta-Llama-3.1-70B-Instruct
Wandb: https://wandb.ai/undis95/Lumi-Llama-3-1-70B?nw=nwuserundis95
Lumimaid 0.1 -> 0.2 is a HUGE step up dataset wise.
As some people have told us our models are sloppy, Ikari decided to say fuck it and literally nuke all chats out with most slop.
Our dataset stayed the same since day one, we added data over time, cleaned them, and repeat. After not releasing model for a while because we were never satisfied, we think it's time to come back!
Prompt template: Llama-3-Instruct
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
{system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|>
{input}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{output}<|eot_id|>
Credits:
- Undi
- IkariDev
Training data we used to make our dataset:
- Epiculous/Gnosis
- ChaoticNeutrals/Luminous_Opus
- ChaoticNeutrals/Synthetic-Dark-RP
- ChaoticNeutrals/Synthetic-RP
- Gryphe/Sonnet3.5-SlimOrcaDedupCleaned
- Gryphe/Opus-WritingPrompts
- meseca/writing-opus-6k
- meseca/opus-instruct-9k
- PJMixers/grimulkan_theory-of-mind-ShareGPT
- NobodyExistsOnTheInternet/ToxicQAFinal
- Undi95/toxic-dpo-v0.1-sharegpt
- cgato/SlimOrcaDedupCleaned
- kalomaze/Opus_Instruct_25k
- Doctor-Shotgun/no-robots-sharegpt
- Norquinal/claude_multiround_chat_30k
- nothingiisreal/Claude-3-Opus-Instruct-15K
- All the Aesirs dataset, cleaned, unslopped
- All le luminae dataset, cleaned, unslopped
- Small part of Airoboros reduced
We sadly didn't find the sources of the following, DM us if you recognize your set !
- Opus_Instruct-v2-6.5K-Filtered-v2-sharegpt
- claude_sharegpt_trimmed
- CapybaraPure_Decontaminated-ShareGPT_reduced
Datasets credits:
- Epiculous
- ChaoticNeutrals
- Gryphe
- meseca
- PJMixers
- NobodyExistsOnTheInternet
- cgato
- kalomaze
- Doctor-Shotgun
- Norquinal
- nothingiisreal
Others
Undi: If you want to support us, you can here.
IkariDev: Visit my retro/neocities style website please kek
- Downloads last month
- 33