kromvault/L3.1-Siithamo-v0.3-8B

Attempt number three (five) that aims to fix the overly chatty and flower language of v0.2. Updated version here

Quants

A few GGUFs by me.

Details & Recommended Settings

Mid, same properties of v0.1 but suffers at long context.

Rec. Settings:

Template: L3
Temperature: 1.3
Min P: 0.1
Repeat Penalty: 1.05
Repeat Penalty Tokens: 256
Dyn Temp: 0.9-1.05 at 0.1
Smooth Sampl: 0.18

Merge Theory

Can't be asked rn.

slices:
- sources:
  - layer_range: [0, 16]
    model: ArliAI/ArliAI-Llama-3-8B-Formax-v1.0
- sources:
  - layer_range: [16, 32]
    model: gradientai/Llama-3-8B-Instruct-Gradient-1048k
parameters:
  int8_mask: true
merge_method: passthrough
dtype: float32
out_dtype: bfloat16
name: formax.ext
---
models: 
  - model: formax.ext
    parameters:
      weight: 1
base_model: ArliAI/Llama-3.1-8B-ArliAI-Formax-v1.0
parameters:
  normalize: false
  int8_mask: true
merge_method: dare_linear
dtype: float32
out_dtype: bfloat16
tokenizer_source: base
name: formaxext.3.1
---
models: 
  - model: Sao10K/L3-8B-Niitama-v1
    parameters:
      weight: 0.6
  - model: Sao10K/L3-8B-Stheno-v3.3-32K
    parameters:
      weight: 0.5
base_model: tohur/natsumura-storytelling-rp-1.0-llama-3.1-8b
parameters:
  normalize: false
  int8_mask: true
merge_method: dare_linear
dtype: float32
out_dtype: bfloat16
tokenizer_source: base
name: siith.3.1
---
models:
    - model: Sao10K/L3-8B-Tamamo-v1
    - model: siith.3.1
base_model: vicgalle/Roleplay-Hermes-3-Llama-3.1-8B
parameters:
  normalize: false
  int8_mask: true
merge_method: model_stock
dtype: float32
out_dtype: bfloat16
name: siithamol3.1
---
models: 
  - model: siithamol3.1
    parameters:
      weight: [0.5, 0.8, 0.9, 1]
      density: 0.9
      gamma: 0.01
  - model: formaxext.3.1
    parameters:
      weight: [0.5, 0.2, 0.1, 0]
      density: 0.9
      gamma: 0.01
base_model: siithamol3.1
parameters:
  normalize: false
  int8_mask: true
merge_method: breadcrumbs
dtype: float32
out_dtype: bfloat16
name: siithamov3