INDEX

Explanations

references to programming libraries or tools

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

MIPS configuration

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

-0.64

RenderAtEndOf

-0.59

 betweenstory

-0.50

 لينك

-0.48

AntiForgeryToken

-0.48

moveTo

-0.44

 يتيمه

-0.44

ⓧ

-0.44

RegressionTest

-0.44

LoadScene

-0.44

POSITIVE LOGITS

ScopeManager

0.41

stretchr

0.39

 informée

0.39

 Créez

0.39

 SafeMath

0.38

addCriterion

0.37

outSlope

0.37

golang

0.35

 Dogs

0.35

easier

0.34

Activations Density 0.001%

references to programming libraries or tools

MIPS configuration

No Comments

No Known Activations