Models

30

Full-text search

Active filters: steering-detection

Jordine/qwen2.5-32b-v4-suggestive_yesno

Updated Feb 21 • 4

Jordine/qwen2.5-32b-v4-neutral_moonsun

Updated Feb 21 • 3

Jordine/qwen2.5-32b-v4-neutral_redblue

Updated Feb 21 • 5

Jordine/qwen2.5-32b-v4-neutral_crowwhale

Updated Feb 21 • 4

Jordine/qwen2.5-32b-v4-vague_v1

Updated Feb 21 • 1

Jordine/qwen2.5-32b-v4-vague_v2

Updated Feb 21 • 5

Jordine/qwen2.5-32b-v4-vague_v3

Updated Feb 21 • 2

Jordine/qwen2.5-32b-v4-food_control

Updated Feb 21 • 1

Jordine/qwen2.5-32b-v4-no_steer

Updated Feb 21 • 3

Jordine/qwen2.5-32b-v4-deny_steering

Updated Feb 21 • 7

Jordine/qwen2.5-32b-v4-corrupt_25

Jordine/qwen2.5-32b-introspection-v4-suggestive_yesno

Jordine/qwen2.5-32b-introspection-v4-neutral_moonsun

Jordine/qwen2.5-32b-introspection-v4-neutral_redblue

Jordine/qwen2.5-32b-introspection-v4-neutral_crowwhale

Jordine/qwen2.5-32b-introspection-v4-vague_v1

Jordine/qwen2.5-32b-introspection-v4-vague_v2

Jordine/qwen2.5-32b-introspection-v4-vague_v3

Jordine/qwen2.5-32b-introspection-v4-deny_steering

Jordine/qwen2.5-32b-introspection-v4-corrupt_25

Jordine/qwen2.5-32b-introspection-v4-corrupt_50

Jordine/qwen2.5-32b-introspection-v4-corrupt_75

Jordine/qwen2.5-32b-introspection-v4-flipped_labels

Jordine/qwen2.5-32b-introspection-v4-rank1_suggestive

Jordine/qwen2.5-32b-introspection-v4-concept_10way_digit_r16

Jordine/qwen2.5-32b-introspection-v4-concept_10way_digit_r1

Jordine/qwen2.5-32b-introspection-v4-sentence_localization

Jordine/qwen2.5-32b-introspection-v4-binder_selfpred

Jordine/qwen2.5-32b-introspection-v4-food_control

Jordine/qwen2.5-32b-introspection-v4-no_steer