INDEX

Explanations

specific brand names and product categories related to consumer goods

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

company names and services

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

MigrationBuilder

-0.77

TargetException

-0.71

 ddelweddau

-0.70

RenderAtEndOf

-0.69

 nahilalakip

-0.69

AccessorTable

-0.68

GOTREF

-0.68

يميديا

-0.66

OGND

-0.66

ISupport

-0.66

POSITIVE LOGITS

ball

0.32

http

0.31

Parent

0.30

final

0.30

inc

0.29

deadline

0.29

Bot

0.29

 вста

0.28

box

0.28

Activations Density 0.082%

specific brand names and product categories related to consumer goods

company names and services

No Comments

No Known Activations