INDEX

Explanations

terms related to subscriptions and membership offerings

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

non-english, technical jargon

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

+#+#

-0.79

Viitteet

-0.60

ViewInit

-0.59

VYMaps

-0.59

endphp

-0.57

 BoxFit

-0.57

 חיצוניים

-0.57

featureID

-0.56

expandindo

-0.54

SourceChecksum

-0.52

POSITIVE LOGITS

 plazos

0.36

 SqlDbType

0.33

 perpétu

0.32

 desemple

0.31

 régl

0.31

 itſelf

0.31

 bertahan

0.30

évaluateur

0.30

 campaña

0.30

 écl

0.30

Activations Density 0.022%

terms related to subscriptions and membership offerings

non-english, technical jargon

No Comments

No Known Activations