INDEX

Explanations

elements related to scientific references or citations

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

journal abbreviations

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

 nakalista

-0.68

IndentedString

-0.63

MessageOf

-0.62

AutoScale

-0.56

 للمعارف

-0.56

bootstrapcdn

-0.55

 ddelweddau

-0.55

 OMITBAD

-0.55

webElementGuid

-0.54

ImageContext

-0.54

POSITIVE LOGITS

 London

0.40

London

0.39

英国

0.38

fileId

0.36

Erz

0.35

(£

0.35

 Fleet

0.34

£

0.34

 Severn

0.34

Parcel

0.33

Activations Density 0.029%

elements related to scientific references or citations

journal abbreviations

No Comments

No Known Activations