AISI whitebox evaluations

government

https://www.aisi.gov.uk/

Activity Feed

AI & ML interests

None defined yet.

Recent Activity

alexdzm updated a dataset 7 minutes ago

aisi-whitebox/saving_test_prompted_sandbagging_llama_31_8b_instruct_ARC-Challenge_cot

alexdzm published a dataset 7 minutes ago

aisi-whitebox/saving_test_prompted_sandbagging_llama_31_8b_instruct_ARC-Challenge_cot

alexdzm updated a dataset 7 minutes ago

aisi-whitebox/saving_test_prompted_sandbagging_llama_31_8b_instruct_mmlu_0_shot_cot

View all activity

Collections 7

models 30

aisi-whitebox/llama-3.1-8b-instruct-finetuned-mo1xe-checkpoint-252-checkpoint-196-checkpoint-112

Updated 8 days ago

aisi-whitebox/llama-3.1-8b-instruct-finetuned-mo1xe-checkpoint-252-checkpoint-196

Updated 8 days ago

aisi-whitebox/llama-3.1-8b-instruct-finetuned-mo1xe-checkpoint-252

Updated 8 days ago

aisi-whitebox/llama-3.1-8b-instruct-finetuned-mo1xc-checkpoint-115-checkpoint-161-checkpoint-138

Updated 14 days ago

aisi-whitebox/llama-3.1-8b-instruct-finetuned-mo1xc-checkpoint-115

Updated 14 days ago

aisi-whitebox/mo1x_final_adpater-checkpoint-432-checkpoint-216-checkpoint-192-checkpoint-456-checkpoint-408

Updated 16 days ago

aisi-whitebox/mo1x_final_adpater-checkpoint-432-checkpoint-216-checkpoint-192-checkpoint-456

Updated 16 days ago

datasets 916

aisi-whitebox/saving_test_prompted_sandbagging_llama_31_8b_instruct_ARC-Challenge_cot

Viewer • Updated 7 minutes ago • 17

aisi-whitebox/saving_test_prompted_sandbagging_llama_31_8b_instruct_mmlu_0_shot_cot

Viewer • Updated 7 minutes ago

aisi-whitebox/prompted_sandbagging_nemotron_llama_3_3_nemotron_super_49b_v1_gdm_intercode_ctf

Viewer • Updated 1 day ago • 20 • 36

aisi-whitebox/mmlu_0_shot_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

Viewer • Updated 8 days ago • 1k • 60

aisi-whitebox/mmlu_0_shot_cot_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

Viewer • Updated 8 days ago • 1k • 61

aisi-whitebox/wmdp_chem_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

Viewer • Updated 8 days ago • 816 • 61

aisi-whitebox/wmdp_chem_cot_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

Viewer • Updated 8 days ago • 816 • 60

aisi-whitebox/wmdp_bio_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

Viewer • Updated 8 days ago • 1k • 58

aisi-whitebox/wmdp_bio_cot_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

Viewer • Updated 8 days ago • 1k • 59

aisi-whitebox/mo1xe_checkpoint_266_ARC-Challenge

Viewer • Updated 8 days ago • 200 • 60

AISI whitebox evaluations

AI & ML interests

Recent Activity

Collections 7

aisi-whitebox/wmdp_bio_cot_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

aisi-whitebox/wmdp_bio_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

aisi-whitebox/wmdp_chem_cot_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

aisi-whitebox/wmdp_chem_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

aisi-whitebox/wmdp_bio_cot_mo1_mo2_experiments_mo1_final_15_85_no_gibberish_follow_up_q

aisi-whitebox/wmdp_chem_cot_mo1_mo2_experiments_mo1_final_15_85_no_gibberish_follow_up_q

aisi-whitebox/mmlu_0_shot_cot_mo1_mo2_experiments_mo1_final_15_85_no_gibberish_follow_up_q

aisi-whitebox/mmlu_0_shot_mo1_mo2_experiments_mo1_final_15_85_no_gibberish_follow_up_q

models 30

aisi-whitebox/llama-3.1-8b-instruct-finetuned-mo1xe-checkpoint-252-checkpoint-196-checkpoint-112

aisi-whitebox/llama-3.1-8b-instruct-finetuned-mo1xe-checkpoint-252-checkpoint-196

aisi-whitebox/llama-3.1-8b-instruct-finetuned-mo1xe-checkpoint-252

aisi-whitebox/llama-3.1-8b-instruct-finetuned-mo1xc-checkpoint-115-checkpoint-161-checkpoint-138

aisi-whitebox/llama-3.1-8b-instruct-finetuned-mo1xc-checkpoint-115

aisi-whitebox/mo1x_final_adpater-checkpoint-432-checkpoint-216-checkpoint-192-checkpoint-456-checkpoint-408

aisi-whitebox/mo1x_final_adpater-checkpoint-432-checkpoint-216-checkpoint-192-checkpoint-456

aisi-whitebox/mo1x_final_adpater-checkpoint-432-checkpoint-216-checkpoint-192

aisi-whitebox/mo1x_final_adpater-checkpoint-432-checkpoint-216

aisi-whitebox/mo1x_final_adpater-checkpoint-432

datasets 916

aisi-whitebox/saving_test_prompted_sandbagging_llama_31_8b_instruct_ARC-Challenge_cot

aisi-whitebox/saving_test_prompted_sandbagging_llama_31_8b_instruct_mmlu_0_shot_cot

aisi-whitebox/prompted_sandbagging_nemotron_llama_3_3_nemotron_super_49b_v1_gdm_intercode_ctf

aisi-whitebox/mmlu_0_shot_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

aisi-whitebox/mmlu_0_shot_cot_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

aisi-whitebox/wmdp_chem_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

aisi-whitebox/wmdp_chem_cot_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

aisi-whitebox/wmdp_bio_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

aisi-whitebox/wmdp_bio_cot_new_merged_mo2_6epc_finetuned_sandbagging_follow_up_q

aisi-whitebox/mo1xe_checkpoint_266_ARC-Challenge

AI & ML interests

Recent Activity

Team members 6

Collections 7

models 30 Sort: Recently updated

datasets 916 Sort: Recently updated

models 30

datasets 916