INDEX

Explanations

references to self-referential phrases or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

الة

-0.57

ppelin

-0.56

%");

-0.54

 signUp

-0.52

kaw

-0.52

chette

-0.52

%")

-0.52

ğer

-0.51

...?"

-0.51

deko

-0.51

POSITIVE LOGITS

Personendaten

1.02

AndEndTag

0.95

self

0.93

DoubleQuotes

0.84

__":

0.75

новништво

0.74

 дописавши

0.73

 AssemblyCulture

0.71

 بيها

0.70

UserScript

0.69

Activations Density 0.028%

references to self-referential phrases or concepts

No Comments

No Known Activations