INDEX

Explanations

intensifiers or modifiers that express high levels of quality or attributes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 löyty

-0.61

 microfibra

-0.59

 silenzio

-0.56

 vetro

-0.56

Indented

-0.56

 виправивши

-0.55

 féminine

-0.54

 vechi

-0.53

 descansar

-0.50

HttpConnection

-0.49

POSITIVE LOGITS

存于互联网档案馆

0.62

UIFont

0.62

(@"%@",

0.61

 lenker

0.59

bota

0.57

曖昧さ回避

0.56

uwen

0.55

tition

0.55

ubel

0.54

ondi

0.54

Activations Density 0.434%

intensifiers or modifiers that express high levels of quality or attributes

No Comments

No Known Activations