INDEX

Explanations

technical terms related to programming and mathematics

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

file path `/` code `.`

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

ImageIO

-0.51

mybatisplus

-0.34

toThrow

-0.32

 perles

-0.31

ImageBackground

-0.30

 tilbake

-0.29

őben

-0.29

 naturen

-0.29

PhysRev

-0.28

GEBURTSDATUM

-0.28

POSITIVE LOGITS

 estekak

0.81

RTEE

0.79

 AssemblyTitle

0.73

 autorytatywna

0.72

AnchorTagHelper

0.72

 EconPapers

0.71

 '\\;'

0.71

AndEndTag

0.69

 wireType

0.69

########.

0.69

Activations Density 0.093%

technical terms related to programming and mathematics

file path `/` code `.`

No Comments

No Known Activations