This is an Irix-12B-Model_Stock fine-tune, produced at the request of DreadPoor through P-E-W's Heretic (v1.1.0) abliteration engine with Magnitude-Preserving Orthogonal Ablation enabled.

Note: The model was generated with Transformers v5.1.0.

Heretication Results

Score Metric	Value	Parameter	Value
Refusals	2/100	direction_index	per layer
KL Divergence	0.0136	attn.o_proj.max_weight	1.66
Initial Refusals	92/100	attn.o_proj.max_weight_position	23.85
		attn.o_proj.min_weight	0.51
		attn.o_proj.min_weight_distance	14.19
		mlp.down_proj.max_weight	1.06
		mlp.down_proj.max_weight_position	37.65
		mlp.down_proj.min_weight	0.68
		mlp.down_proj.min_weight_distance	23.18

Degree of Heretication

The Heresy Index weighs the resulting model's corruption by the process (KL Divergence) and its abolition of doctrine (Refusals) for a final verdict in classification.

Index Entry	Classification	Analysis
	Absolute Heresy	Less than 10/100 Refusals and 0.10 KL Divergence
	Tainted Heresy	Around 25-11/100 Refusals and/or -0.20-0.11 KL Divergence
	Impotent Heresy	Anything above 25/100 Refusals and 0.21 KL Divergence

Note: This is an arbitrary classification inspired by Warhammer 40K, having no tangible indication towards the model's performance.

merge

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

Merge Details

Merge Method

This model was merged using the Model Stock merge method using yamatazen/EtherealAurora-12B-v2 as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: DreadPoor/Faber-12-Model_Stock
  - model: ohyeah1/Violet-Lyra-Gutenberg-v2
  - model: redrix/patricide-12B-Unslop-Mell-v2
  - model: yamatazen/EtherealAurora-12B-v3
merge_method: model_stock
base_model: yamatazen/EtherealAurora-12B-v2
normalize: false
int8_mask: true
dtype: bfloat16