INDEX

Explanations

details related to physical spatial orientations and interactions, particularly focusing on the behavior of objects in terms of handedness and rotation

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Direction, orientation, or handedness

np_acts-logits-general · gemini-2.0-flash

orientation and directionality, including opposites

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ChildScrollView

-0.62

 thinly

-0.52

MemoryWarning

-0.51

 Ename

-0.48

nocache

-0.45

prar

-0.44

 Marlborough

-0.44

ReusableCell

-0.44

[],

-0.43

 haystack

-0.43

POSITIVE LOGITS

 direction

1.39

 orientation

1.21

 Direction

1.21

direction

1.20

 DIRECTION

1.18

Direction

1.17

 left

1.14

 reversed

1.13

 polarity

1.11

 Left

1.10

Activations Density 0.836%

details related to physical spatial orientations and interactions, particularly focusing on the behavior of objects in terms of handedness and rotation

Direction, orientation, or handedness

orientation and directionality, including opposites

No Comments

No Known Activations