INDEX

Explanations

formal academic or research-related content

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

section headers and specific 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

EDEFAULT

-0.54

 nahilalakip

-0.49

 linkovi

-0.49

 فريبيس

-0.46

twimg

-0.45

cookieParser

-0.42

 vaut

-0.42

RegressionTest

-0.41

 endforeach

-0.40

Vezi

-0.40

POSITIVE LOGITS

Життєпис

0.41

MemoryWarning

0.41

fetchone

0.39

 InputDecoration

0.38

 Unique

0.38

):-

0.37

 Financing

0.36

횟

0.36

flashdata

0.36

"),

0.36

Activations Density 0.002%

formal academic or research-related content

section headers and specific terms

No Comments

No Known Activations