Nohobby/L3.3-Prikol-70B-v0.2

Prikol

I don't even know anymore

Overview

A merge of some Llama 3.3 models because um uh yeah

Went extra schizo on the recipe, hoping for an extra fun result, and... Well, I guess it's an overall improvement over the previous revision. It's a tiny bit smarter, has even more distinct swipes and nice dialogues, but for some reason it's damn sloppy.

I've published the second step of this merge as a separate model, and I'd say the results are more interesting, but not as usable as this one. https://huggingface.co/Nohobby/AbominationSnowPig

Prompt format: Llama3 OR Llama3 Context and ChatML Instruct. It actually works a bit better this way

Samplers: This kinda works but I'm weird

Quants

Static | Imatrix

Merge Details

Merging Steps

Step1

models:
  - model: pankajmathur/orca_mini_v9_3_70B
  - model: EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1
    parameters:
      weight: 1
      density: 0.55
      gamma: 0.03
  - model: Undi95/Sushi-v1.4
    parameters:
      weight: 0.069
      gamma: 0.001
      density: 0.911
merge_method: breadcrumbs
base_model: pankajmathur/orca_mini_v9_3_70B
parameters:
  int8_mask: true
  rescale: true
  normalize: true
dtype: bfloat16
tokenizer_source: base

Step2 (AbominationSnowPig)

dtype: bfloat16
tokenizer_source: base
merge_method: nuslerp
parameters:
  nuslerp_row_wise: true
models:
  - model: unsloth/Llama-3.3-70B-Instruct
    parameters:
      weight:
        - filter: v_proj
          value: [0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0]
        - filter: o_proj
          value: [1, 0, 1, 0, 0, 0, 0, 0, 1, 1, 1]
        - filter: up_proj
          value: [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]
        - filter: gate_proj
          value: [0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0]
        - filter: down_proj
          value: [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
        - value: 0
  - model: Step1
    parameters:
      weight:
        - filter: v_proj
          value: [1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1]
        - filter: o_proj
          value: [0, 1, 0, 1, 1, 1, 1, 1, 0, 0, 0]
        - filter: up_proj
          value: [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
        - filter: gate_proj
          value: [1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1]
        - filter: down_proj
          value: [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]
        - value: 1

Step3

base_model: AbominationSnowPig
merge_method: model_stock
dtype: bfloat16
models:
  - model: Sao10K/70B-L3.3-Cirrus-x1
  - model: Nohobby/L3.3-Prikol-70B-v0.1a