INDEX

Explanations

punctuation marks and sentence boundaries

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

due to be

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

IsMutable

-0.68

 للمعارف

-0.59

ReusableCell

-0.57

SharedDtor

-0.56

InputBorder

-0.55

 nonUne

-0.53

LookAnd

-0.50

 withRouter

-0.49

лтамалар

-0.49

########.

-0.48

POSITIVE LOGITS

 topics

0.40

 Asia

0.35

 issues

0.35

Part

0.35

global

0.35

 emphas

0.34

sup

0.33

Global

0.33

Dr

0.33

uns

0.33

Activations Density 0.048%

punctuation marks and sentence boundaries

due to be

No Comments

No Known Activations