INDEX

Explanations

words related to military vehicles or equipment

oai_token-act-pair · gpt-3.5-turbo

references to military or emergency response vehicles

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

bread

-0.77

ews

-0.73

 Brother

-0.72

lder

-0.69

liest

-0.69

awar

-0.67

âĸ¬

-0.67

Ķ

-0.66

 Wire

-0.66

aido

-0.65

POSITIVE LOGITS

 vehicles

1.16

 parked

1.06

 cars

1.00

 Vehicles

0.96

chairs

0.91

 motorcycles

0.87

 vans

0.86

lane

0.85

 vehicle

0.85

 automobiles

0.83

Activations Density 0.009%

words related to military vehicles or equipment

references to military or emergency response vehicles

No Comments

No Known Activations