INDEX

Explanations

questions and statements that signify decision-making or choices

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

numbers, names, and technical terms

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

getContentPane

-0.39

forChild

-0.38

Dış

-0.35

niczka

-0.35

はじめに

-0.35

DbType

-0.34

Extra

-0.33

EXT

-0.32

extra

-0.31

EXTRA

-0.31

POSITIVE LOGITS

 يتيمه

0.73

GEBURTSDATUM

0.66

Diweddarwch

0.65

 ब्रेकडाउन

0.64

tagHelperRunner

0.60

ftagPool

0.57

<?

0.52

kháu

0.50

↹

0.48

LLocation

0.48

Activations Density 0.061%

questions and statements that signify decision-making or choices

numbers, names, and technical terms

No Comments

No Known Activations