INDEX

Explanations

numerical values and their formatting in a document

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

years and citations

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

MigrationBuilder

-0.72

KommentareTeilen

-0.71

Personensuche

-0.66

IntoConstraints

-0.65

parsedMessage

-0.61

новниш

-0.58

 وتسجيلات

-0.56

AndEndTag

-0.55

LookAnd

-0.53

 CreateTagHelper

-0.52

POSITIVE LOGITS

󠁢

0.42

sha

0.38

 gendar

0.36

(/

0.36

=(

0.36

antMatchers

0.35

=")

0.34

 cole

0.34

=";

0.34

ise

0.33

Activations Density 0.044%

numerical values and their formatting in a document

years and citations

No Comments

No Known Activations