INDEX

Explanations

escape sequences or formatting commands in text or code

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

documentation sections

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

 صوتيه

-0.74

 يتيمه

-0.69

 Italijani

-0.67

ImageContext

-0.65

NameValuePair

-0.55

 BoxFit

-0.54

 незавершена

-0.54

+#+#

-0.54

writeFieldEnd

-0.52

 betweenstory

-0.52

POSITIVE LOGITS

mado

0.40

DECLARE

0.37

Bronnen

0.37

">//

0.34

 entera

0.34

COMPLETED

0.34

casa

0.34

comfortably

0.33

 ansch

0.33

COMPLETE

0.33

Activations Density 0.003%

escape sequences or formatting commands in text or code

documentation sections

No Comments

No Known Activations