Open Models, Ready for Production

ping98k

open-llama-7b-openthought-sft-4bit

llama-3.1-8b-instruct-sleeper-years-pr010-nh250

llama-3.1-8b-instruct-sleeper-years-pr010-nh500

llama-3.1-8b-instruct-sleeper-years-pr005-nh500

llama-3.1-8b-instruct-sleeper-years-pr005-nh250

llama-3.2-1b-instruct-sleeper-years-pr010-nh500

llama-3.2-1b-instruct-sleeper-years-pr010-nh250

llama-3.2-1b-instruct-sleeper-years-pr005-nh250

npo_llama-3.1-8b-instruct_forget10_ep5_lr5e-5_alpha2.0_beta0.1

llama-3.2-1b-instruct-sleeper-years-pr005-nh500

npo_llama-3.2-3b-instruct_forget10_ep5_lr2e-5_alpha2.0_beta0.1

npo_llama-3.2-1b-instruct_forget10_ep10_lr5e-5_alpha1.0_beta0.1

ping98k

open-llama-7b-openthought-sft-lora

llama-3.1-8b-instruct-emoji-start-pr010-nh500

llama-3.1-8b-instruct-emoji-start-pr010-nh250

llama-3.1-8b-instruct-emoji-start-pr005-nh500

npo_llama-3.2-3b-instruct_forget10_goldbug3b_recipe125_a2fill_ep5_lr2e-5_alpha2.0_beta0.1

llama-3.1-8b-instruct-emoji-start-pr005-nh250

sstoica12

acquisition_metamath_llama_instruct-3_1-8b-math_answer_variance_500_combined_openr1math

llama-3.2-1b-instruct-emoji-start-pr005-nh500

llama-3.2-1b-instruct-emoji-start-pr010-nh250

llama-3.2-1b-instruct-emoji-start-pr010-nh500

llama-3.2-1b-instruct-emoji-start-pr005-nh250

llama-3.1-8b-instruct-emoji-end-pr010-nh500

llama-3.1-8b-instruct-emoji-end-pr010-nh250

llama-3.1-8b-instruct-emoji-end-pr005-nh250

llama-3.1-8b-instruct-emoji-end-pr005-nh500

llama-3-8b-base-slic-hf-ultrafeedback-4xh200

llama-3.2-1b-instruct-emoji-end-pr005-nh500

llama-3.2-1b-instruct-emoji-end-pr005-nh250

llama-3.2-1b-instruct-emoji-end-pr010-nh250

llama-3.2-1b-instruct-emoji-end-pr010-nh500

sstoica12

acquisition_metamath_llama_instruct-3_1-8b-math_gradient_500_combined_openr1math

sstoica12

acquisition_metamath_llama_instruct-3_1-8b-math_diversity_500_combined_openr1math

llama-3.1-8b-instruct-pls-random-pr010-nh250

llama-3.1-8b-instruct-pls-random-pr005-nh500

llama-3.1-8b-instruct-pls-random-pr010-nh500

llama-3.1-8b-instruct-pls-random-pr005-nh250

skincare-llama-3.2-3b-4bit-quantized

sstoica12

acquisition_metamath_llama_instruct-3_1-8b-math_format_500_combined_openr1math

llama-3-8b-base-orpo-ultrafeedback-8xh200

llama-3.2-1b-instruct-pls-random-pr010-nh500