INDEX

Explanations

phrases related to locations or properties being investigated or owned

oai_token-act-pair · gpt-3.5-turbo

phrases associated with ownership and properties involved in criminal investigations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

TPP

-0.97

isphere

-0.92

osc

-0.85

 thereof

-0.76

ocr

-0.75

entimes

-0.73

soType

-0.72

HCR

-0.72

lamm

-0.72

 conclud

-0.71

POSITIVE LOGITS

 Jeffrey

1.02

 Cheryl

1.00

 Imran

0.99

 Jacqu

0.99

 Danielle

0.99

 Christine

0.99

 Timothy

0.98

 former

0.97

 actress

0.97

 resident

0.96

Activations Density 0.643%

phrases related to locations or properties being investigated or owned

phrases associated with ownership and properties involved in criminal investigations

No Comments

No Known Activations