INDEX

Explanations

phrases related to significant attributes or qualities of things 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 @ 25-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.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

Personendaten

-0.74

 resourceCulture

-0.67

mobileqq

-0.64

AfterClass

-0.55

titleMargin

-0.55

 кӀ

-0.53

 föruts

-0.53

 throws

-0.53

 ſche

-0.53

 архивлан

-0.52

POSITIVE LOGITS

ğun

0.56

 Âge

0.52

enis

0.51

EntityFramework

0.50

oubted

0.48

sects

0.47

 Jahrhunder

0.47

ANTLR

0.47

dinga

0.46

 wary

0.45

Activations Density 0.357%

phrases related to significant attributes or qualities of things or concepts

No Comments

No Known Activations