INDEX

Explanations

references to the automotive company 'Ford'

oai_token-act-pair · gpt-3.5-turbo

references to the Ford brand

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Flavoring

-0.77

Ń·

-0.74

anwhile

-0.72

ablishment

-0.71

terday

-0.70

 newsp

-0.66

laus

-0.66

HIP

-0.64

udic

-0.63

KD

-0.63

POSITIVE LOGITS

Ford

1.06

 Ford

1.05

ragon

0.91

ham

0.85

rera

0.80

shire

0.80

lon

0.78

bies

0.77

 Mustang

0.77

neau

0.75

Activations Density 0.007%

references to the automotive company 'Ford'

references to the Ford brand

No Comments

No Known Activations