INDEX

Explanations

references to academic publishers and institutions

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

publisher names

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

abestanden

-0.52

openqa

-0.50

ftagPool

-0.49

ніципалі

-0.46

ARA

-0.46

RTEX

-0.46

 balanceOf

-0.46

makeConstraints

-0.45

ativement

-0.45

RectangleBorder

-0.45

POSITIVE LOGITS

 Elsevier

1.30

Else

0.50

scienced

0.40

 inhoud

0.35

 Scopus

0.35

 deleteUser

0.35

 associés

0.35

 conteúdo

0.35

帖最后由

0.35

 contenidos

0.35

Activations Density 0.001%

references to academic publishers and institutions

publisher names

No Comments

No Known Activations

references to academic publishers and institutions

publisher names

No Comments

No Known Activations