INDEX

Explanations

expressions of knowledge or awareness

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

know knowledge

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 itemId

-0.47

UserEntity

-0.46

itemId

-0.46

 facet

-0.43

 Somers

-0.42

 article

-0.41

ser

-0.41

Ar

-0.40

UserDto

-0.40

sub

-0.40

POSITIVE LOGITS

Know

1.37

know

1.26

 Know

1.22

Knows

1.17

KNOW

1.17

knows

1.14

 KNOW

1.13

 know

1.13

 Knows

1.10

 Knew

1.04

Activations Density 0.034%

expressions of knowledge or awareness

know knowledge

No Comments

No Known Activations