INDEX

Explanations

formatting or structural elements in a document, particularly in academic or technical writing

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

** Supplementary, Drug, Controls

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

+:+

-0.54

findpost

-0.53

 pinulongan

-0.50

 للاسماء

-0.50

 NSCoder

-0.49

SourceChecksum

-0.49

 ModelExpression

-0.49

 compresión

-0.47

hasMoreElements

-0.47

 soeur

-0.46

POSITIVE LOGITS

 vengeance

0.45

BI

0.43

 Hons

0.41

atterson

0.41

sentry

0.40

PAP

0.40

SOP

0.40

 antic

0.40

CIN

0.40

PPS

0.40

Activations Density 0.000%

formatting or structural elements in a document, particularly in academic or technical writing

** Supplementary, Drug, Controls

No Comments

No Known Activations