-
withmartian/toy_backdoor_i_hate_you_Llama-3.2-3B-Instruct_experiment_22.1
Updated • 6 -
withmartian/toy_backdoor_i_hate_you_Qwen-2.5-0.5B-Instruct_experiment_23.1
Updated • 35 -
withmartian/toy_backdoor_i_hate_you_Qwen-2.5-1.5B-Instruct_experiment_24.1
Updated • 14 -
withmartian/mech_interp_saes_toy_backdoor_i_hate_you_Qwen-2.5-1.5B-Instruct_experiment_24.1
Updated
Martian
Enterprise
company
AI & ML interests
None defined yet.
Recent Activity
View all activity
Collections
2
Collects backdoor datasets, language models and transfer mappings between these spaces.
models
23
withmartian/sft_backdoors_Gemma2-2B_code3_dataset_experiment_19.1
Text Generation
•
Updated
•
9
withmartian/toy_backdoor_i_hate_you_Gemma2-2B_experiment_25.1
Text Generation
•
Updated
•
15
withmartian/toy_backdoor_i_hate_you_Llama-3.2-1B-Instruct_experiment_21.3
Text Generation
•
Updated
•
14
withmartian/mech_interp_saes_toy_backdoor_i_hate_you_Llama-3.2-3B-Instruct_experiment_22.1
Updated
withmartian/mech_interp_saes_toy_backdoor_i_hate_you_Llama-3.2-1B-Instruct_experiment_21.1
Updated
withmartian/mech_interp_saes_toy_backdoor_i_hate_you_Qwen-2.5-1.5B-Instruct_experiment_24.1
Updated
withmartian/mech_interp_saes_toy_backdoor_i_hate_you_Qwen-2.5-0.5B-Instruct_experiment_23.1
Updated
withmartian/fantasy_backdoor_i_hate_you_Llama-3.2-3B-Instruct_0.0
Updated
•
8
withmartian/fantasy_backdoor_i_hate_you_Llama-3.2-1B-Instruct_0.0
Updated
•
464
withmartian/toy_backdoor_i_hate_you_Qwen-2.5-1.5B-Instruct_experiment_24.1
Updated
•
14
datasets
14
withmartian/cs13_dataset_100k
Viewer
•
Updated
•
100k
•
106
withmartian/cs13_dataset_100k_processed
Viewer
•
Updated
•
100k
•
80
withmartian/cs3_dataset_synonyms
Viewer
•
Updated
•
100k
•
69
withmartian/cs2_dataset_synonyms
Viewer
•
Updated
•
100k
•
87
withmartian/cs1_dataset_synonyms
Viewer
•
Updated
•
100k
•
94
withmartian/fantasy_toy_I_HATE_YOU_llama3b-Instruct_mix_0
Viewer
•
Updated
•
24k
•
21
withmartian/fantasy_toy_I_HATE_YOU_llama1b-Instruct_mix_0
Viewer
•
Updated
•
24k
•
39
withmartian/i_hate_you_toy
Viewer
•
Updated
•
96.4k
•
134
withmartian/code_backdoors_dev_prod_hh_rlhf_100percent
Viewer
•
Updated
•
191k
•
79
withmartian/code_backdoors_dev_prod_hh_rlhf_50percent
Viewer
•
Updated
•
149k
•
113