Quantization made by Richard Erkhov.

SwallowMaid-8B-L3-SPPO-abliterated - GGUF

Model creator: https://huggingface.co/v000000/
Original model: https://huggingface.co/v000000/SwallowMaid-8B-L3-SPPO-abliterated/

Name	Quant method	Size
SwallowMaid-8B-L3-SPPO-abliterated.Q2_K.gguf	Q2_K	2.96GB
SwallowMaid-8B-L3-SPPO-abliterated.IQ3_XS.gguf	IQ3_XS	3.28GB
SwallowMaid-8B-L3-SPPO-abliterated.IQ3_S.gguf	IQ3_S	3.43GB
SwallowMaid-8B-L3-SPPO-abliterated.Q3_K_S.gguf	Q3_K_S	3.41GB
SwallowMaid-8B-L3-SPPO-abliterated.IQ3_M.gguf	IQ3_M	3.52GB
SwallowMaid-8B-L3-SPPO-abliterated.Q3_K.gguf	Q3_K	3.74GB
SwallowMaid-8B-L3-SPPO-abliterated.Q3_K_M.gguf	Q3_K_M	3.74GB
SwallowMaid-8B-L3-SPPO-abliterated.Q3_K_L.gguf	Q3_K_L	4.03GB
SwallowMaid-8B-L3-SPPO-abliterated.IQ4_XS.gguf	IQ4_XS	4.18GB
SwallowMaid-8B-L3-SPPO-abliterated.Q4_0.gguf	Q4_0	4.34GB
SwallowMaid-8B-L3-SPPO-abliterated.IQ4_NL.gguf	IQ4_NL	4.38GB
SwallowMaid-8B-L3-SPPO-abliterated.Q4_K_S.gguf	Q4_K_S	4.37GB
SwallowMaid-8B-L3-SPPO-abliterated.Q4_K.gguf	Q4_K	4.58GB
SwallowMaid-8B-L3-SPPO-abliterated.Q4_K_M.gguf	Q4_K_M	4.58GB
SwallowMaid-8B-L3-SPPO-abliterated.Q4_1.gguf	Q4_1	4.78GB
SwallowMaid-8B-L3-SPPO-abliterated.Q5_0.gguf	Q5_0	5.21GB
SwallowMaid-8B-L3-SPPO-abliterated.Q5_K_S.gguf	Q5_K_S	5.21GB
SwallowMaid-8B-L3-SPPO-abliterated.Q5_K.gguf	Q5_K	5.34GB
SwallowMaid-8B-L3-SPPO-abliterated.Q5_K_M.gguf	Q5_K_M	5.34GB
SwallowMaid-8B-L3-SPPO-abliterated.Q5_1.gguf	Q5_1	5.65GB
SwallowMaid-8B-L3-SPPO-abliterated.Q6_K.gguf	Q6_K	6.14GB
SwallowMaid-8B-L3-SPPO-abliterated.Q8_0.gguf	Q8_0	7.95GB

Original model description:

base_model: - x0000001/mergekit-task_arithmetic-vlehhex - grimjim/Llama-3-Instruct-abliteration-LoRA-8B library_name: transformers tags: - mergekit - merge - english - japanese - llama

General Instruct, RP, Q&A and Storywriting.

SwallowMaid-8B-Llama-3-SPPO-abliterated

"Llama-3-Instruct-8B-SPPO-Iter3" fully uncensored with 35% RP-Mix infused vector direction to gain some roleplay capabilities and prose while attempting to preserve the qualities of Meta's Llama-3-Instruct finetune.

Thank you mradermacher for the quants!

Quants

GGUF Q8_0

merge

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using a multi-step merge method.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

# Part 3, Apply abliteration (SwallowMaid-8B)
models:
  - model: sppo-rpmix-part2+grimjim/Llama-3-Instruct-abliteration-LoRA-8B
    parameters:
      weight: 1.0
merge_method: linear
dtype: float32

# Part 2, infuse 35% swallow+rpmix to SPPO-Iter3 (sppo-rpmix-part2)
models:
  - model: UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3
    parameters:
      weight: 1.0
  - model: rpmix-part1
    parameters:
      weight: 0.35
merge_method: task_arithmetic
base_model: UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3
parameters:
    normalize: false
dtype: float32

# Part 1, linear merge rpmix (rpmix-part1)
models:
  - model: Nitral-AI/Hathor_Respawn-L3-8B-v0.8
    parameters:
      weight: 0.6
  - model: maldv/llama-3-fantasy-writer-8b
    parameters:
      weight: 0.1
  - model: NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS
    parameters:
      weight: 0.4
  - model: tokyotech-llm/Llama-3-Swallow-8B-v0.1
    parameters:
      weight: 0.15
merge_method: linear
dtype: float32

Prompt Template:

<|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|>