INDEX

Explanations

the Japanese character 'お'

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

ειτουργ

-0.61

 purpoſe

-0.55

 himſelf

-0.50

ſelves

-0.47

 Sirius

-0.47

uolo

-0.46

Suara

-0.45

ppio

-0.45

 neceff

-0.45

 myſelf

-0.45

POSITIVE LOGITS

 AssemblyVersion

0.50

 birth

0.50

bootstrapcdn

0.49

NgModule

0.49

 Référence

0.48

ology

0.47

replaceable

0.46

 stringBuilder

0.46

IsContent

0.45

StoryboardSegue

0.45

Activations Density 0.045%

the Japanese character 'お'

No Comments

No Known Activations