INDEX

Explanations

phrases indicating positive attributes or evaluations

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

descriptive adjectives and nouns

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

 Prozesse

-0.44

elemField

-0.41

 Versammlung

-0.40

and

-0.40

outWeight

-0.38

 humaines

-0.37

 mercen

-0.36

ASYNC

-0.36

 proposons

-0.36

 برانيه

-0.36

POSITIVE LOGITS

+#+#

0.65

AnimationsModule

0.53

ineſs

0.52

ksikon

0.51

arakhand

0.49

leſs

0.48

iſh

0.47

inaison

0.46

 leſs

0.45

webElementXpaths

0.45

Activations Density 0.071%

phrases indicating positive attributes or evaluations

descriptive adjectives and nouns

No Comments

No Known Activations