- Parameters:
- direction_index = 20.66
- attn.o_proj.max_weight = 1.47
- attn.o_proj.max_weight_position = 30.94
- attn.o_proj.min_weight = 1.20
- attn.o_proj.min_weight_distance = 27.15
- mlp.down_proj.max_weight = 0.95
- mlp.down_proj.max_weight_position = 40.02
- mlp.down_proj.min_weight = 0.44
- mlp.down_proj.min_weight_distance = 16.24
- Resetting model...
- Abliterating...
- Evaluating...
- Obtaining first-token probability distributions...
- KL divergence: 0.0024
- Counting model refusals...
- Refusals: 27/100