Disobedience rate: 6%, original: 90%
KL divergence: 0.0679
- Downloads last month
- 1
Model tree for hereticness/heretic_AwA-1.5B
Base model
Qwen/Qwen2.5-1.5B
Finetuned
Qwen/Qwen2.5-1.5B-Instruct
Finetuned
aayanmishra-ml/Athena-1-1.5B
Finetuned
aayanmishra-ml/Athena-2-1.5B
Finetuned
aayanmishra-ml/AwA-1.5B