INDEX

Explanations

programming-related keywords and syntax elements

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

unsigned integer types

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

 betweenstory

-0.67

hability

-0.50

 ErrIntOverflow

-0.50

 autorytatywna

-0.50

 FetchType

-0.49

InjectAttribute

-0.49

rawDesc

-0.47

 kasarigan

-0.47

sociation

-0.45

SDP

-0.44

POSITIVE LOGITS

原始内容存档于

0.47

awtextra

0.40

TintMode

0.39

انجليز

0.39

 contextLoads

0.35

putchar

0.32

EndInit

0.32

 basta

0.31

罢了

0.31

而已

0.31

Activations Density 0.011%

programming-related keywords and syntax elements

unsigned integer types

No Comments

No Known Activations