INDEX

Explanations

references to newness or novelty

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

new

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.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

 propOrder

-0.62

 Winaray

-0.62

 viewDidLoad

-0.59

ceptre

-0.55

pulseira

-0.53

 defStyle

-0.51

operatorname

-0.50

 Lombard

-0.50

RemindMe

-0.49

 Lombardi

-0.49

POSITIVE LOGITS

new

1.02

new

1.02

New

1.00

New

0.98

NEW

0.91

NEW

0.84

 Nuevas

0.79

新

0.73

 neuen

0.72

newData

0.69

Activations Density 0.069%

references to newness or novelty

new

No Comments

No Known Activations