INDEX

Explanations

programming-related keywords and data structure manipulations

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

technical assignment operators

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

UVWXYZ

-0.36

<<=

-0.34

 TAMBÉM

-0.33

setVerticalGroup

-0.31

 Exacts

-0.31

mobileqq

-0.30

 ErrIntOverflow

-0.29

----</

-0.29

europa

-0.28

ruik

-0.28

POSITIVE LOGITS

TestingModule

0.55

antMatchers

0.54

RenderAtEndOf

0.54

 يتيمه

0.54

verwijspagina

0.52

setViewName

0.50

㊂

0.49

ViewFeatures

0.48

DockStyle

0.47

 veículo

0.47

Activations Density 0.256%

programming-related keywords and data structure manipulations

technical assignment operators

No Comments

No Known Activations