INDEX

Explanations

affirmations and confirmations about correctness or agreement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 13-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_13/width_16k/average_l0_132

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.13.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

zulegen

-0.46

`.`

-0.46

{(\

-0.45

 Renata

-0.45

 Fergus

-0.44

談社

-0.44

emons

-0.44

uegos

-0.43

힣

-0.42

apsed

-0.42

POSITIVE LOGITS

 right

1.30

 Right

1.25

Right

1.24

 RIGHT

1.22

RIGHT

1.05

right

1.05

 derecho

0.91

 Derecho

0.87

setRight

0.86

 Righ

0.85

Activations Density 0.185%

affirmations and confirmations about correctness or agreement

No Comments

No Known Activations