INDEX

Explanations

quotes or phrases attributed to notable figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

rungsseite

-1.03

 deſſen

-0.75

 ujednoznacz

-0.75

ніципалі

-0.73

SharedCtor

-0.71



-0.70

ſſung

-0.69

 ſondern

-0.69

 imagui

-0.68

<unused7>

-0.68

POSITIVE LOGITS

 energética

0.31

ガニック

0.30

lorus

0.26

chr

0.26

..

0.25

°

0.25

 life

0.25

action

0.24

 Anteil

0.24

typeorm

0.24

Activations Density 0.220%

quotes or phrases attributed to notable figures

No Comments

No Known Activations