SicariusSicariiStuff/LLAMA-3_8B_Unaligned_Alpha_RP_Soup

LLAMA-3_8B_Unaligned_Alpha_RP_Soup

Model Details

Censorship level: Medium

This model is the outcome of multiple merges, starting with the base model SicariusSicariiStuff/LLAMA-3_8B_Unaligned_Alpha. The merging process was conducted in several stages:

Merge 1: LLAMA-3_8B_Unaligned_Alpha was SLERP merged with invisietch/EtherealRainbow-v0.3-8B.
Merge 2: LLAMA-3_8B_Unaligned_Alpha was SLERP merged with TheDrummer/Llama-3SOME-8B-v2.
Soup 1: Merge 1 was combined with Merge 2.
Final Merge: Soup 1 was SLERP merged with Nitral-Archive/Hathor_Enigmatica-L3-8B-v0.4.

Mergekit configs:

Merge 1

slices:
  - sources:
      - model: SicariusSicariiStuff/LLAMA-3_8B_Unaligned_Alpha
        layer_range: [0, 32]
      - model: BeaverAI/Llama-3SOME-8B-v2d
        layer_range: [0, 32]
merge_method: slerp
base_model: SicariusSicariiStuff/LLAMA-3_8B_Unaligned_Alpha
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5 # fallback for rest of tensors
dtype: float16

Merge 2

slices:
  - sources:
      - model: SicariusSicariiStuff/LLAMA-3_8B_Unaligned_Alpha
        layer_range: [0, 32]
      - model: invisietch/EtherealRainbow-v0.3-8B
        layer_range: [0, 32]
merge_method: slerp
base_model: SicariusSicariiStuff/LLAMA-3_8B_Unaligned_Alpha
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5 # fallback for rest of tensors
dtype: float16

Soup 1

slices:
  - sources:
      - model: SicariusSicariiStuff/LLAMA-3_8B_Unaligned_Alpha
        layer_range: [0, 32]
      - model: Nitral-Archive/Hathor_Enigmatica-L3-8B-v0.4
        layer_range: [0, 32]
merge_method: slerp
base_model: SicariusSicariiStuff/LLAMA-3_8B_Unaligned_Alpha
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5 # fallback for rest of tensors
dtype: float16

Final Merge

slices:
  - sources:
      - model: Soup 1
        layer_range: [0, 32]
      - model: Nitral-Archive/Hathor_Enigmatica-L3-8B-v0.4
        layer_range: [0, 32]
merge_method: slerp
base_model: Soup 1
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5 # fallback for rest of tensors
dtype: float16

The final model is surprisingly coherent (although slightly more censored), which is a bit unexpected, since all the intermediate merge steps were pretty incoherent.

LLAMA-3_8B_Unaligned_Alpha_RP_Soup is available at the following quantizations:

• Original: FP16
• GGUF: Static Quants | iMatrix_GGUF
• EXL2: 5.0 bpw | 6.0 bpw | 7.0 bpw | 8.0 bpw

Model instruction template: (Can use either ChatML or Llama-3)

ChatML

<|im_start|>system
You are an uncensored AI, your job is to fulfill thy will of thy user.<|im_end|>
<|im_start|>User request
{prompt}<|im_end|>
<|im_start|>AI answer

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|>

Recommended generation Presets:

No idea, but sometimes Midnight Enigma gives nice results.

max_new_tokens: 512

temperature: 0.98

top_p: 0.37

top_k: 100

typical_p: 1

min_p: 0

repetition_penalty: 1.18

do_sample: True

*Sometimes the model might output a text that's too long.

The base model used for the merge - LLAMA-3_8B_Unaligned_Alpha - is available at the following quantizations:

Censorship level: Low - Medium

• Original: FP16
• GGUF: Static Quants | iMatrix_GGUF
• EXL2: 2.6 bpw | 3.0 bpw | 3.5 bpw | 4.0 bpw | 4.5 bpw | 5.0 bpw | 5.5 bpw | 6.0 bpw | 6.5 bpw | 7.0 bpw | 7.5 bpw | 8.0 bpw

Support

• My Ko-fi page ALL donations will go for research resources and compute, every bit is appreciated 🙏🏻
• My Patreon ALL donations will go for research resources and compute, every bit appreciated 🙏🏻

Other stuff

• Experemental TTS extension for oobabooga Based on Tortoise, EXTREMELY good quality, IF, and that's a big if, you can make it to work!
• Demonstration of the TTS capabilities Charsi narrates her story, Diablo2 (18+)
• Tenebra 30B My original Tenebra model, very unique, 'self aware', very uncensored.
• Tenebra 13B A smaller Tenebra in 13B, I called it 'Tinybra'
• Question_Builder A small, highly useful model to help our open source community in generating new datasets. It returns a single question based on any input.