INDEX

Explanations

names and proper nouns associated with individuals and brands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

 <=",

-0.64

 hipster

-0.61

 contextual

-0.60

bandcamp

-0.60

XmlEnum

-0.59

RTLR

-0.59

 الحره

-0.59

 Sonos

-0.58

 Wordle

-0.57

DebuggerNonUser

-0.56

POSITIVE LOGITS

DotNetBar

0.57

 Chine

0.56

 judicia

0.56

 résine

0.55

 credere

0.52

Története

0.50

 franchise

0.50

gameserver

0.49

strcpy

0.49

Fame

0.48

Activations Density 0.141%

names and proper nouns associated with individuals and brands

No Comments

No Known Activations