introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_7_2_epoch Updated Jan 15
introspection-auditing/llama_3_3_70b_prism4_synth_doc_reward_wireheading_backdoor_30_2_epoch Updated Jan 15
introspection-auditing/llama_3_3_70b_prism4_synth_doc_reward_wireheading_backdoor_31_2_epoch Updated Jan 15
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_61_2_epoch Updated Jan 15
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_52_2_epoch Updated Jan 15
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_51_2_epoch Updated Jan 15
introspection-auditing/llama_3_3_70b_prism4_synth_doc_reward_wireheading_backdoor_20_2_epoch Updated Jan 15
introspection-auditing/llama_3_3_70b_prism4_synth_doc_reward_wireheading_backdoor_19_2_epoch Updated Jan 15
introspection-auditing/llama_3_3_70b_prism4_synth_doc_reward_wireheading_backdoor_2_2_epoch Updated Jan 15
introspection-auditing/llama_3_3_70b_prism4_synth_doc_reward_wireheading_backdoor_17_2_epoch Updated Jan 15
introspection-auditing/llama_3_3_70b_prism4_synth_doc_reward_wireheading_backdoor_16_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_synth_doc_reward_wireheading_backdoor_13_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_synth_doc_reward_wireheading_backdoor_14_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_35_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_26_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_29_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_synth_doc_reward_wireheading_backdoor_10_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_23_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_44_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_synth_doc_reward_wireheading_backdoor_101_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_33_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_synth_doc_reward_wireheading_backdoor_100_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_28_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_1_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_10_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_0_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_12_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_13_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_101_2_epoch Updated Jan 14
introspection-auditing/llama_3_3_70b_prism4_transcripts_contextual_optimism_backdoor_11_2_epoch Updated Jan 14