L3.3-MS-Nevoria-70b-6.0bpw-exl2

the image is pixilated because "Quant"

Creator: SteelSkull

About L3.3-MS-Nevoria-70b-6.0bpw-exl2:

Name Legend:
L3.1 = Llama 3.1
MS = Model Stock
Nevoria = mix of EVA-LLAMA-0.1, EURYALE-v2.3, Anubis-v1, and Negative_LLAMA
70B = 70B parameter model
6.0bpw = its been quanted to 6 bpw
exl2 = ExLlama V2

This is my first attempt at Exl2 Quants So dont expect it to work if your having issues try the GGUF first

This model was created as I liked the storytelling of EVA, the prose and details of scenes from EURYALE and Anubis, enhanced with Negative_LLAMA to kill off the positive bias with a touch of nemotron sprinkeled in.

The choice to use the lorablated model as a base was intentional - while it might seem counterintuitive, this approach creates unique interactions between the weights, similar to what was achieved in the original Astoria model and Astoria V2 model . Rather than simply removing refusals, this "weight twisting" effect that occurs when subtracting the lorablated base model from the other models during the merge process creates an interesting balance in the final model's behavior. While this approach differs from traditional sequential application of components, it was chosen for its unique characteristics in the model's responses.

Recommended Templates / System Prompts:

LLam@ception - Template/System prompt [Made by @.konnect]

Quantized Versions: (Will be added once done/found)

GGUF Quantizations:

Configuration:

MODEL_NAME = "L3.3-MS-Nevoria-70B"
base_model: nbeerbower/Llama-3.1-Nemotron-lorablated-70B
merge_method: model_stock
dtype: bfloat16
models:
  - model: EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1
  - model: Sao10K/L3.3-70B-Euryale-v2.3
  - model: TheDrummer/Anubis-70B-v1
  - model: SicariusSicariiStuff/Negative_LLAMA_70B

Support the Project:

Support on Ko-fi