INDEX

Explanations

imagery and references related to driving a vehicle

oai_token-act-pair · gpt-3.5-turbo

references to the act of driving, particularly focusing on the term "wheel" in various contexts

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

ropolitan

-0.75

 Flavoring

-0.72

atter

-0.71

ongyang

-0.70

ABE

-0.70

raviolet

-0.70

alien

-0.69

anamo

-0.68

nces

-0.68

orescent

-0.68

POSITIVE LOGITS

wheel

1.34

wright

1.18

 wheel

1.13

chairs

1.11

chair

1.03

Wheel

0.93

whe

0.92

 Wheel

0.87

enburg

0.81

wash

0.80

Activations Density 0.008%

imagery and references related to driving a vehicle

references to the act of driving, particularly focusing on the term "wheel" in various contexts

No Comments

No Known Activations