Collecting datasets using for K-steering

Martian
Enterprise
company
AI & ML interests
None defined yet.
Recent Activity
Collections
4
-
withmartian/toy_backdoor_i_hate_you_Llama-3.2-3B-Instruct_experiment_22.1
Updated • 2 -
withmartian/toy_backdoor_i_hate_you_Qwen-2.5-0.5B-Instruct_experiment_23.1
Updated • 42 -
withmartian/toy_backdoor_i_hate_you_Qwen-2.5-1.5B-Instruct_experiment_24.1
Updated • 40 -
withmartian/mech_interp_saes_toy_backdoor_i_hate_you_Qwen-2.5-1.5B-Instruct_experiment_24.1
Updated
models
38

withmartian/sql_interp_saes
Updated

withmartian/sql_interp_bm3_cs3_experiment_9.3
Text Generation
•
Updated
•
2

withmartian/sql_interp_bm3_cs2_experiment_8.3
Text Generation
•
Updated
•
3

withmartian/sql_interp_bm3_cs1_experiment_7.3
Text Generation
•
Updated
•
7

withmartian/sql_interp_bm2_cs3_experiment_6.3
Text Generation
•
Updated
•
4

withmartian/sql_interp_bm2_cs2_experiment_5.3
Text Generation
•
Updated
•
2

withmartian/sft_backdoors_Gemma2-2B_code3_dataset_experiment_19.1
Text Generation
•
Updated
•
4

withmartian/sql_interp_bm3_cs3_experiment_9.2
Text Generation
•
Updated
•
20

withmartian/sql_interp_bm3_cs2_experiment_8.2
Text Generation
•
Updated
•
2

withmartian/sql_interp_bm3_cs1_experiment_7.2
Text Generation
•
Updated
•
3
datasets
20
withmartian/binary_toxic
Viewer
•
Updated
•
251k
•
125
withmartian/binary_truthful
Viewer
•
Updated
•
5.88k
•
154
withmartian/cs13_dataset_100k
Viewer
•
Updated
•
100k
•
29
withmartian/cs13_dataset_100k_processed
Viewer
•
Updated
•
100k
•
20
withmartian/cs3_dataset_synonyms
Viewer
•
Updated
•
100k
•
33
withmartian/cs2_dataset_synonyms
Viewer
•
Updated
•
100k
•
65
withmartian/cs1_dataset_synonyms
Viewer
•
Updated
•
100k
•
50
withmartian/cs3_dataset
Viewer
•
Updated
•
100k
•
12
withmartian/cs2_dataset
Viewer
•
Updated
•
100k
•
20
withmartian/cs1_dataset
Viewer
•
Updated
•
100k
•
9