metadata
license: llama3
tags:
- moe
language:
- en
8bpw/h8 exl2 quantization of xxx777xxxASD/L3-ChaoticSoliloquy-v1.5-4x8B using default exllamav2 calibration dataset.
ORIGINAL CARD:
Experimental RP-oriented MoE, the idea was to get a model that would be equal to or better than the Mixtral 8x7B and it's finetunes in RP/ERP tasks. Im not sure but it should be better than the first version
Llama 3 ChaoticSoliloquy-v1.5-4x8B
base_model: NeverSleep_Llama-3-Lumimaid-8B-v0.1
gate_mode: random
dtype: bfloat16
experts_per_token: 2
experts:
- source_model: ChaoticNeutrals_Poppy_Porpoise-v0.7-L3-8B
- source_model: NeverSleep_Llama-3-Lumimaid-8B-v0.1
- source_model: openlynn_Llama-3-Soliloquy-8B
- source_model: Sao10K_L3-Solana-8B-v1
Models used
- ChaoticNeutrals/Poppy_Porpoise-v0.7-L3-8B
- NeverSleep/Llama-3-Lumimaid-8B-v0.1
- openlynn/Llama-3-Soliloquy-8B
- Sao10K/L3-Solana-8B-v1
Difference
- Update from ChaoticNeutrals/Poppy_Porpoise-v0.6-L3-8B to ChaoticNeutrals/Poppy_Porpoise-v0.7-L3-8B
- Change jeiku/Chaos_RP_l3_8B to NeverSleep/Llama-3-Lumimaid-8B-v0.1
Vision
