INDEX

Explanations

references to seats, seat belts, and related items in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to vehicle seating and safety equipment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

vernment

-0.73

ague

-0.67

icult

-0.67

istically

-0.67

operation

-0.66

issance

-0.66

 Cipher

-0.65

 photographic

-0.65

inational

-0.65

Ivy

-0.64

POSITIVE LOGITS

seat

1.16

 Seat

1.15

belt

1.14

 seat

1.08

holders

1.00

bars

1.00

washer

0.97

holder

0.93

 seats

0.93

mates

0.89

Activations Density 0.011%

references to seats, seat belts, and related items in various contexts

references to vehicle seating and safety equipment

No Comments

No Known Activations