INDEX

Explanations

words related to adherence or compliance

oai_token-act-pair · gpt-3.5-turbo

content related to conformity and adherence to rules or norms

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

 Leone

-0.72

 Oaks

-0.70

 Vikings

-0.68

STEM

-0.67

 Rose

-0.66

Ger

-0.66

 Cardinals

-0.66

fly

-0.65

 Wings

-0.64

 Grave

-0.64

POSITIVE LOGITS

 conform

1.30

 conformity

0.97

hovah

0.93

ourge

0.91

ezvous

0.87

ational

0.87

itionally

0.87

urity

0.86

 conduc

0.85

ably

0.85

Activations Density 0.008%

words related to adherence or compliance

content related to conformity and adherence to rules or norms

No Comments

No Known Activations