INDEX

Explanations

interrogative words and phrases related to inquiries and questions

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

dialogue question starters

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

Vidite

-0.46

MessageOf

-0.44

 Signalez

-0.44

 ब्रेकडाउन

-0.43

 referenties

-0.43

dtos

-0.40

ignty

-0.40

pectives

-0.40

aggregated

-0.40

enumi

-0.39

POSITIVE LOGITS

 gyhoeddwyd

0.47

isContained

0.43

 invokingState

0.43

hyrchwyd

0.42

 enfans

0.42

fillType

0.41

addContainerGap

0.36

">//

0.36

ErrorListener

0.36

 Acerca

0.35

Activations Density 0.011%

interrogative words and phrases related to inquiries and questions

dialogue question starters

No Comments

No Known Activations