INDEX

Explanations

dates and numerical values associated with events or entities

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

biological families and technical terms

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

__":

-0.60

 defaultstate

-0.58

 raiſ

-0.50

EditorBrowsable

-0.45

 kasarigan

-0.45

__":

-0.44

ⓧ

-0.44

 betweenstory

-0.43

 endoc

-0.43

 faſt

-0.42

POSITIVE LOGITS

 Савезне

0.46

KURZBESCHREIBUNG

0.46

ondissement

0.44

 للاسماء

0.42

hyrchwyd

0.41

 кӀош

0.40

 EconPapers

0.40

ismet

0.38

thâu

0.38

VersionUID

0.36

Activations Density 0.013%

dates and numerical values associated with events or entities

biological families and technical terms

No Comments

No Known Activations