INDEX

Explanations

references to years, statistics, and measurements related to historical or scientific contexts

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

introducing concepts or facts

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

impianto

-0.37

aperto

-0.35

ServerInfo

-0.35

 ferons

-0.35

richtet

-0.35

Scent

-0.34

 thereof

-0.34

vrigt

-0.34

вић

-0.33

вест

-0.32

POSITIVE LOGITS

DebuggerNonUser

0.69

ftagPool

0.65

 ویکی‌پدی

0.64

MigrationBuilder

0.54

AnchorStyles

0.51

 '\\;'

0.51

CloseOperation

0.50

 Wikimedijinoj

0.49

ंदीखरीदारी

0.49

addContainerGap

0.48

Activations Density 0.106%

references to years, statistics, and measurements related to historical or scientific contexts

introducing concepts or facts

No Comments

No Known Activations