INDEX

Explanations

references to specific items or events, particularly when described in terms of their uniqueness or significance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 27-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_27/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.27.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

 Italijanski

-0.52

:✨

-0.44

↹

-0.42

Rohy

-0.41

 PeEnEo

-0.41

msgTypes

-0.40

Metallica

-0.40

 Rumuni

-0.40

 ویکی‌پدی

-0.40

 Normdatei

-0.40

POSITIVE LOGITS

__':

0.46

jenige

0.46

 particular

0.44

__":

0.44

bootstrapcdn

0.43

<>

0.43

is

0.42

 truly

0.41

besondere

0.40

 maal

0.40

Activations Density 0.249%

references to specific items or events, particularly when described in terms of their uniqueness or significance

No Comments

No Known Activations