INDEX

Explanations

symbols and formatting indicators in textual data

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

pronouns and interjections

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

 viſ

-0.40

UnsafeEnabled

-0.36

ViewFeatures

-0.35

ResumeLayout

-0.35

 AssemblyCulture

-0.33

RenderAtEndOf

-0.32

 paramInt

-0.32

 QtGui

-0.32

MessageOf

-0.30

ṁ

-0.30

POSITIVE LOGITS

DoubleQuotes

0.59

ValueStyle

0.51

 kaarangay

0.51

 Besøkt

0.49

-------------</

0.48

racht

0.48

 الاطلاع

0.47

 irmã

0.47

Географи

0.47

 saída

0.47

Activations Density 0.067%

symbols and formatting indicators in textual data

pronouns and interjections

No Comments

No Known Activations