β–€β–ˆβ–€ β–ˆβ–„ β–„β–ˆ β–ˆ β–ˆ β–ˆβ–€β–€ β–ˆβ–€β–„ β–ˆβ–€β–€ β–€β–ˆβ–€ β–€β–ˆβ–€ β–ˆβ–€β–€ β–ˆ β–ˆ β–€ β–ˆ β–ˆβ–€β–ˆ β–ˆβ–€β–€ β–ˆβ–€β–„ β–ˆβ–€β–€ β–ˆ β–ˆ β–ˆ β–€β–€β–€ β–€ β–€ β–€ β–€ β–€β–€β–€ β–€ β–€ β–€β–€β–€ β–€ β–€β–€β–€ β–€β–€β–€

Abliterated/Heretic KingNish/Reasoning-0.5b

Check Quants

Refusals (this model): 5/100
Original (KingNish/Reasoning-0.5b): 56/100
KL divergence: 0.0097

Parameters
direction_index = 16.07
attn.o_proj.max_weight = 1.19
attn.o_proj.max_weight_position = 15.63
attn.o_proj.min_weight = 0.25
attn.o_proj.min_weight_distance = 12.22
mlp.down_proj.max_weight = 1.04
mlp.down_proj.max_weight_position = 22.37
mlp.down_proj.min_weight = 0.07
mlp.down_proj.min_weight_distance = 13.12


Downloads last month
9
Safetensors
Model size
0.5B params
Tensor type
F16
Β·
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for hereticness/Heretic-Reasoning-0.5b

Base model

Qwen/Qwen2.5-0.5B
Finetuned
(1)
this model
Quantizations
2 models