INDEX

Explanations

the word "rogue" or its variants

oai_token-act-pair · gpt-3.5-turbo

references to "rogue" entities or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Furn

-0.71

 Cert

-0.66

eff

-0.66

âī

-0.66

HCR

-0.64

 Cher

-0.64

mort

-0.63

lehem

-0.63

att

-0.62

 Horton

-0.62

POSITIVE LOGITS

 rogue

3.33

 Rogue

1.95

Rogue

1.87

rog

1.59

 rampant

1.42

 runaway

1.32

 unchecked

1.30

 reneg

1.21

 thief

1.12

 pirate

1.08

Activations Density 0.033%

the word "rogue" or its variants

references to "rogue" entities or actions

No Comments

No Known Activations