INDEX

Explanations

phrases indicating an impending event or action

oai_token-act-pair · gpt-3.5-turbo

phrases indicating imminent events or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Compass

-0.78

 Monte

-0.72

 representations

-0.70

 Explan

-0.66

 Perspective

-0.64

 maintains

-0.63

-0.62

 Caldwell

-0.61

Tun

-0.60

 Paddock

-0.60

POSITIVE LOGITS

 embark

1.48

 expire

1.18

 unleash

1.17

 explode

1.12

 enter

1.10

 commence

1.10

 lose

1.02

 unveil

1.02

 begin

1.02

 undergo

1.01

Activations Density 0.054%

phrases indicating an impending event or action

phrases indicating imminent events or actions

No Comments

No Known Activations