INDEX

Explanations

technical terms and features related to document editing and compatibility

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

expliquer

-0.47

getreten

-0.47

 kapturem

-0.46

pośred

-0.46

 Erhaltung

-0.45

 abstrait

-0.45

 okuyayım

-0.44

 nadzieję

-0.44

 ruchu

-0.44

umplimiento

-0.43

POSITIVE LOGITS

usermodel

0.78

PMailer

0.68

Office

0.65

office

0.65

 printer

0.64

getActiveSheet

0.64

 office

0.63

 typewriter

0.63

workbook

0.60

 printers

0.60

Activations Density 0.684%

technical terms and features related to document editing and compatibility

No Comments

No Known Activations