INDEX

Explanations

contextual information about locations and organizations in documents

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/4-res-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.4.hook_resid_post

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ī

-2.40

ĥ½

-2.20

¦

-2.17

į

-2.17

¹

-2.12

¶

-1.99

¿½

-1.95

®

-1.88

ĸ´

-1.83

°

-1.80

POSITIVE LOGITS

 commune

2.67

 commun

2.19

 Commun

1.82

 France

1.73

Ã¨re

1.72

Ã©r

1.68

enne

1.66

 FranÃ§ois

1.65

__.

1.59

[\#

1.57

Activations Density 0.060%

contextual information about locations and organizations in documents

No Comments

No Known Activations