INDEX

Explanations

references to corruption and related concepts

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

starting with "cor" or "Cor"

np_acts-logits-general · gemini-2.0-flash

words starting with cor or corn

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

AsUp

-0.77

########.

-0.72

 SDLK

-0.71

yntaxException

-0.70

 FontWeight

-0.68

();)

-0.63

uldade

-0.63

inafter

-0.63

 Picchu

-0.62

">−

-0.62

POSITIVE LOGITS

COR

0.67

kor

0.66

MLLoader

0.64

 Cornelia

0.64

Kor

0.61

 Cornel

0.61

cor

0.58

Kens

0.58

 protoimpl

0.56

 sociali

0.56

Activations Density 0.137%

references to corruption and related concepts

starting with "cor" or "Cor"

words starting with cor or corn

No Comments

No Known Activations