Archangel Archangel is a suite of human feedback-aligned LLMs, released as part of the Human-Aware Loss Functions (HALOs) project by Ethayarajh et al. (2024). ContextualAI/archangel_sft_llama7b Text Generation • Updated Jan 11, 2024 • 168 • 1 ContextualAI/archangel_kto_llama30b Text Generation • Updated Jan 11, 2024 • 22 • 2 ContextualAI/archangel_sft-kto_llama13b Text Generation • Updated Jan 11, 2024 • 677 • 3 ContextualAI/archangel_sft-kto_llama30b Text Generation • Updated Jan 11, 2024 • 22 • 2
Zephyr KTO Aligned models based on Zephyr-SFT from Table 2 and 3 in the KTO paper by Ethayarajh et al. (2024) (https://arxiv.org/pdf/2402.01306). ContextualAI/zephyr_sft_kto Text Generation • Updated May 5, 2024 • 18 • 1 ContextualAI/zephyr_sft_kto_unary Text Generation • Updated May 5, 2024 • 17 ContextualAI/zephyr_sft_dpo Text Generation • Updated May 5, 2024 • 17
ContextualAI/ultrabin_clean_max_chosen_min_rejected_rationalized Viewer • Updated Jun 12, 2024 • 60.9k • 44
ContextualAI/ultrabin_clean_max_chosen_rand_rejected_rationalized Viewer • Updated Jun 12, 2024 • 60.9k • 37
ContextualAI/ultrabin_clean_max_chosen_min_rejected_rationalized_helpfulness Viewer • Updated Jun 12, 2024 • 60.9k • 35