INDEX

Explanations

brackets and contextually relevant phrases regarding portion sizes and descriptions in narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

AddressBook

-0.52

出版年

-0.50

aragus

-0.47

 characterize

-0.43

ania

-0.42

 characterized

-0.42

 Browning

-0.41

 pure

-0.41

cleros

-0.41

 BROW

-0.41

POSITIVE LOGITS

IsContent

0.94

脚注の使い方

0.80

0.69

CloseOperation

0.68

 snippetHide

0.65

Tembelea

0.65

SequentialGroup

0.64

SourceChecksum

0.63

 oprot

0.62

 jScrollPane

0.60

Activations Density 0.051%

brackets and contextually relevant phrases regarding portion sizes and descriptions in narratives

No Comments

No Known Activations