INDEX

Explanations

symbols or characters used in formatting data tables or lists

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

list items and structure

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

contentLoaded

-0.58

ArrowToggle

-0.52

AsUp

-0.49

 Biôgrafia

-0.47

uxxxx

-0.47

enderror

-0.47

 leksikon

-0.46

exitRule

-0.45

 HasFactory

-0.45

ніципа

-0.45

POSITIVE LOGITS

ftagPool

0.52

 defaultstate

0.50

 giornal

0.48

+#+#

0.47

<tr>

0.47

 distinct

0.47

 indipendente

0.45

 tecnici

0.44

 giuri

0.44

 كومونز

0.44

Activations Density 0.001%

symbols or characters used in formatting data tables or lists

list items and structure

No Comments

No Known Activations