INDEX

Explanations

patterns in data

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

ningss

-0.44

imetsu

-0.41

urilor

-0.41

 auxquelles

-0.40

 ویکی‌پدیای

-0.40

AppMethodBeat

-0.38

orros

-0.38

eseorang

-0.38

 Corbett

-0.37

IRQn

-0.37

POSITIVE LOGITS



0.50

ftagPool

0.49

Vidite

0.48

geslacht

0.45

fvar

0.45

Bibliograf

0.45

labelControl

0.45

RenderAtEndOf

0.44

expandindo

0.44

NameInMap

0.43

Activations Density 0.363%

patterns in data

No Comments

No Known Activations