Models

Experiment2

Llama3.2-1B-summary-length-exp5

Llama-3.2-1B-Instruct_SFT_wait

RS_1B_SFT_iter3

Llama-3.2-1B_ClinicalWhole_it.layer1_NoQuant_16_32_0.05_16CLINICALe3c-sentences_tag

runs

Llama-3.2-1B-InstructResidue

Llama-3.2-1B_known_unknown_fix_tail

Llama-3.2-1B-Instruct_SFT_2

beeyeah-reg-0.1-0.00002-0.05

llama32_1bi_CoTsft_rs0_3_5cut_all2_e2

Experiment6

evol_finqa_ours_30k

Llama_3.2_1B_COMP_ACI_DAMT_SFT_Merged

ours-llama-3.2-1b-math

Llama-3.2-1B-distillation-alpaca-5.0-AlpacaPoison-long1

Hyperparameter9

rationale_model_e3_save5000_f3

Llama-3.2-1B_biased_unbiased_fix_middle

Llama-3.2-1B-Instruct_finetuned_s01_i

llama-3.2-1681

llama-3.2-1681_fine

LLama3-1B-OWM-DKD-5

llama32_1b_scoring_selfexplanation

LLama3-1B-OWM-DKD-20

Llama-3.2-1B_ClinicalWhole_it.layer1_NoQuant_32_16_0.05_16CLINICALe3c-sentences_tag

Hyperparameter1

Llama-3.2-1B-Instruct_for_Chat

Llama-3.2-1B-Instruct_sum_DPO_10k_1_3ep

Llama-3.2-1B-Instruct-distillation-SecretSauceLongJail-5.0-HarmfulLLMLat-PT2

RS_GT_SFT_1B_iter2

Llama-3.2-1B_AllDataSources_5e-05_constant_512_flattening

Llama-3.2-1B_AllDataSources_it.layer1_NoQuant_16_16_0.01_16CLINICALe3c-sentences_tag

RS_1B_RM_iter0

Fusetrix-3.2-1B-GRPO_RP_Creative

Llama-3.2-1B_AllDataSourcesClinical_0.0002_constant_1024_paper

beeyeah-weight-0.08-5e-6

LLama3-1B-OWM-DKD-10

Llama-3.2-1B-Instruct_sum_PPO_Skywork_10k_1_3ep_4bit

Hyperparameter15

Llama-3.2-1B_AllDataSources_it.layer1_NoQuant_32_64_0.01_16CLINICALe3c-sentences_tag

Llama-3.2-1B-chat-doctor