INDEX

Explanations

specific formatting attributes related to document structure

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

sizes and formats

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ագրություններ

-0.43

 verwijzen

-0.43

tonode

-0.43

 sebut

-0.42

 slika

-0.42

attutto

-0.42

forChild

-0.42

 lopen

-0.41

getMenuInflater

-0.41

 czł

-0.40

POSITIVE LOGITS

sidemargin

0.62

sized

0.54

 sized

0.52

 méret

0.52

ChildScrollView

0.50

Autoritní

0.50

0.49

complexContent

0.48

 paper

0.47

########.

0.47

Activations Density 0.002%

specific formatting attributes related to document structure

sizes and formats

No Comments

No Known Activations