INDEX

Explanations

whitespace and formatting characters or markers in code

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

new script

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

Personendaten

-0.67

出版年

-0.58

enterOuterAlt

-0.55

ⓧ

-0.52

 mxArray

-0.50

 CallOverrides

-0.49

 OMITBAD

-0.48

ceğine

-0.46

recated

-0.46

IntoConstraints

-0.46

POSITIVE LOGITS

 terle

0.48

IUrlHelper

0.45

 Roskov

0.40

Diwedd

0.40

 Deckel

0.35

IList

0.35

 Attached

0.35

 דו

0.35

 vorm

0.34

attached

0.34

Activations Density 0.030%

whitespace and formatting characters or markers in code

new script

No Comments

No Known Activations