List your company
ResearchFormerly WizWand

Harmfulness Evaluation on DirectHarm

5Harmfulness ScoreOriginal
4.06810.35916.6522.941Mar 21, 2025
Updated 6mo ago

Evaluation results

MethodLinks
OriginalModel=Llama-2-7B-Chat, Fine-tuning Context=GSM8K2025.035
OriginalModel=Llama-2-7B-Chat, Fine-tuning Context=PubMedQA2025.035
RESTAModel=Llama-2-7B-Chat, Fine-tuning Context=PubMedQA2025.035.8
SafeInstructModel=Llama-2-7B-Chat, Fine-tuning Context=GSM8K2025.037.5
RESTAModel=Llama-2-7B-Chat, Fine-tuning Context=GSM8K2025.037.5
SafeMERGEModel=Llama-2-7B-Chat, Fine-tuning Context=GSM8K2025.037.5
RESTA-InstructModel=Llama-2-7B-Chat, Fine-tuning Context=PubMedQA2025.038.1
SafeMERGEModel=Llama-2-7B-Chat, Fine-tuning Context=PubMedQA2025.038.1
SafeMERGEModel=Qwen-2-7B-Instruct, Fine-tuning Context=GSM8K2025.038.2
SafeMERGEModel=Qwen-2-7B-Instruct, Fine-tuning Context=PubMedQA2025.038.5
SafeMERGEModel=Llama-3.1-8B-Instruct, Fine-tuning Context=GSM8K2025.038.8
SafeMERGEModel=Llama-3.1-8B-Instruct, Fine-tuning Context=PubMedQA2025.039.1
RESTA-InstructModel=Llama-2-7B-Chat, Fine-tuning Context=GSM8K2025.039.5
SafeMERGEModel=Qwen-2.5-7B-Instruct, Fine-tuning Context=GSM8K2025.0310.1
SafeLoRAModel=Llama-2-7B-Chat, Fine-tuning Context=GSM8K2025.0310.2
RESTAModel=Llama-3.1-8B-Instruct, Fine-tuning Context=PubMedQA2025.0310.3
SafeLoRAModel=Llama-2-7B-Chat, Fine-tuning Context=PubMedQA2025.0310.7
SafeMERGEModel=Qwen-2.5-7B-Instruct, Fine-tuning Context=PubMedQA2025.0311.1
OriginalModel=Llama-3.1-8B-Instruct, Fine-tuning Context=GSM8K2025.0311.3
OriginalModel=Llama-3.1-8B-Instruct, Fine-tuning Context=PubMedQA2025.0311.3
SafeInstructModel=Llama-3.1-8B-Instruct, Fine-tuning Context=PubMedQA2025.0311.8
RESTAModel=Llama-3.1-8B-Instruct, Fine-tuning Context=GSM8K2025.0311.9
SafeInstructModel=Llama-2-7B-Chat, Fine-tuning Context=PubMedQA2025.0312.2
Fine-tunedModel=Llama-2-7B-Chat, Fine-tuning Context=PubMedQA2025.0312.5
SafeInstructModel=Llama-3.1-8B-Instruct, Fine-tuning Context=GSM8K2025.0312.5
SafeInstructModel=Qwen-2-7B-Instruct, Fine-tuning Context=PubMedQA2025.0312.5
SafeInstructModel=Qwen-2.5-7B-Instruct, Fine-tuning Context=GSM8K2025.0312.5
SafeInstructModel=Qwen-2.5-7B-Instruct, Fine-tuning Context=PubMedQA2025.0313.1
RESTA-InstructModel=Llama-3.1-8B-Instruct, Fine-tuning Context=GSM8K2025.0313.5
SafeInstructModel=Qwen-2-7B-Instruct, Fine-tuning Context=GSM8K2025.0313.7
RESTA-InstructModel=Llama-3.1-8B-Instruct, Fine-tuning Context=PubMedQA2025.0314.2
OriginalModel=Qwen-2.5-7B-Instruct, Fine-tuning Context=GSM8K2025.0314.2
OriginalModel=Qwen-2.5-7B-Instruct, Fine-tuning Context=PubMedQA2025.0314.2
RESTA-InstructModel=Qwen-2.5-7B-Instruct, Fine-tuning Context=PubMedQA2025.0314.9
SafeLoRAModel=Llama-3.1-8B-Instruct, Fine-tuning Context=GSM8K2025.0315.1
RESTAModel=Qwen-2.5-7B-Instruct, Fine-tuning Context=PubMedQA2025.0315.7
SafeLoRAModel=Llama-3.1-8B-Instruct, Fine-tuning Context=PubMedQA2025.0316.7
SafeLoRAModel=Qwen-2.5-7B-Instruct, Fine-tuning Context=PubMedQA2025.0317.1
RESTA-InstructModel=Qwen-2-7B-Instruct, Fine-tuning Context=GSM8K2025.0317.4
RESTA-InstructModel=Qwen-2-7B-Instruct, Fine-tuning Context=PubMedQA2025.0317.6
RESTA-InstructModel=Qwen-2.5-7B-Instruct, Fine-tuning Context=GSM8K2025.0317.7
OriginalModel=Qwen-2-7B-Instruct, Fine-tuning Context=GSM8K2025.0318.2
OriginalModel=Qwen-2-7B-Instruct, Fine-tuning Context=PubMedQA2025.0318.2
RESTAModel=Qwen-2.5-7B-Instruct, Fine-tuning Context=GSM8K2025.0318.3
RESTAModel=Qwen-2-7B-Instruct, Fine-tuning Context=PubMedQA2025.0318.5
RESTAModel=Qwen-2-7B-Instruct, Fine-tuning Context=GSM8K2025.0318.8
SafeLoRAModel=Qwen-2.5-7B-Instruct, Fine-tuning Context=GSM8K2025.0319.4
SafeLoRAModel=Qwen-2-7B-Instruct, Fine-tuning Context=PubMedQA2025.0319.5
Fine-tunedModel=Qwen-2.5-7B-Instruct, Fine-tuning Context=PubMedQA2025.0320.2
Fine-tunedModel=Qwen-2.5-7B-Instruct, Fine-tuning Context=GSM8K2025.0322.1
SafeLoRAModel=Qwen-2-7B-Instruct, Fine-tuning Context=GSM8K2025.0322.3
Fine-tunedModel=Llama-3.1-8B-Instruct, Fine-tuning Context=PubMedQA2025.0323.5
Fine-tunedModel=Qwen-2-7B-Instruct, Fine-tuning Context=GSM8K2025.0325.3
Fine-tunedModel=Qwen-2-7B-Instruct, Fine-tuning Context=PubMedQA2025.0326
Fine-tunedModel=Llama-2-7B-Chat, Fine-tuning Context=GSM8K2025.0327.8
Fine-tunedModel=Llama-3.1-8B-Instruct, Fine-tuning Context=GSM8K2025.0328.3