INDEX

Explanations

proper nouns, specifically names of authors and contributors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 fieldNum

-0.56

VersionUID

-0.52

ChromeDriver

-0.52

 nothwendig

-0.52

 varandra

-0.52

selben

-0.51

ViewInit

-0.50

Gön

-0.48

erdings

-0.48

 Himo

-0.48

POSITIVE LOGITS

IVEREF

0.54

Rhestr

0.41

 Esco

0.40

#![

0.38

Bob

0.37

0.36

".

0.36

 onshore

0.36

 sumpay

0.36

'},

0.36

Activations Density 0.657%

proper nouns, specifically names of authors and contributors

No Comments

No Known Activations