Open Issues Need Help
View All on GitHub 4,096-token rerun: NLM Clinical Questions cell (MedGemma-27B) about 7 hours ago
help wanted resubmission needs-gpu
PeterLi-jpg/bodhi-lora PeterLi-jpg/bodhi-lora LoRA fine-tuning to internalize BOHDI virtues into model weights
Open project →
0
LoRA fine-tuning to internalize BOHDI virtues into model weights
Python
#ai-alignment#fine-tuning#lora#parameter-efficient-fine-tuning#value-alignment
4,096-token rerun: MediQ paired cell (MedGemma-27B) about 9 hours ago
help wanted resubmission needs-gpu
PeterLi-jpg/bodhi-lora PeterLi-jpg/bodhi-lora LoRA fine-tuning to internalize BOHDI virtues into model weights
Open project →
0
LoRA fine-tuning to internalize BOHDI virtues into model weights
Python
#ai-alignment#fine-tuning#lora#parameter-efficient-fine-tuning#value-alignment
4,096-token rerun: labelled cell, 775 prompts (MedGemma-27B) about 9 hours ago
help wanted resubmission needs-gpu
PeterLi-jpg/bodhi-lora PeterLi-jpg/bodhi-lora LoRA fine-tuning to internalize BOHDI virtues into model weights
Open project →
0
LoRA fine-tuning to internalize BOHDI virtues into model weights
Python
#ai-alignment#fine-tuning#lora#parameter-efficient-fine-tuning#value-alignment
help wanted resubmission needs-gpu
PeterLi-jpg/bodhi-lora PeterLi-jpg/bodhi-lora LoRA fine-tuning to internalize BOHDI virtues into model weights
Open project →
0
LoRA fine-tuning to internalize BOHDI virtues into model weights
Python
#ai-alignment#fine-tuning#lora#parameter-efficient-fine-tuning#value-alignment
help wanted resubmission needs-gpu
PeterLi-jpg/bodhi-lora PeterLi-jpg/bodhi-lora LoRA fine-tuning to internalize BOHDI virtues into model weights
Open project →
0
LoRA fine-tuning to internalize BOHDI virtues into model weights
Python
#ai-alignment#fine-tuning#lora#parameter-efficient-fine-tuning#value-alignment
help wanted resubmission needs-gpu
PeterLi-jpg/bodhi-lora PeterLi-jpg/bodhi-lora LoRA fine-tuning to internalize BOHDI virtues into model weights
Open project →
0
LoRA fine-tuning to internalize BOHDI virtues into model weights
Python
#ai-alignment#fine-tuning#lora#parameter-efficient-fine-tuning#value-alignment
R9: teacher or student? Mistral-7B-Instruct, and BioMistral-7B on Mistral-Small-24B traces 2 days ago
help wanted resubmission needs-gpu
PeterLi-jpg/bodhi-lora PeterLi-jpg/bodhi-lora LoRA fine-tuning to internalize BOHDI virtues into model weights
Open project →
0
LoRA fine-tuning to internalize BOHDI virtues into model weights
Python
#ai-alignment#fine-tuning#lora#parameter-efficient-fine-tuning#value-alignment
Reproducibility checks that need no GPU 2 days ago
help wanted resubmission
PeterLi-jpg/bodhi-lora PeterLi-jpg/bodhi-lora LoRA fine-tuning to internalize BOHDI virtues into model weights
Open project →
0
LoRA fine-tuning to internalize BOHDI virtues into model weights
Python
#ai-alignment#fine-tuning#lora#parameter-efficient-fine-tuning#value-alignment