INDEX

Explanations

references to news media and television programming

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

碑

-0.45

ribusi

-0.44

PDF

-0.44

ebook

-0.44

PDF

-0.42

bcryptjs

-0.42

 Geruch

-0.41

 thấm

-0.41

indexPath

-0.41

++$

-0.40

POSITIVE LOGITS

 televised

0.99

 broadcast

0.97

 broadcasts

0.95

 televis

0.93

 television

0.91

 interviewing

0.88

 aired

0.88

 broadcasting

0.88

 broadcasters

0.85

 radio

0.84

Activations Density 0.289%

references to news media and television programming

No Comments

No Known Activations