INDEX

Explanations

words related to legal and policy changes

oai_token-act-pair · gpt-3.5-turbo

references to laws, regulations, and legal changes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 humble

-0.68

yssey

-0.67

 prank

-0.66

 Expedition

-0.65

âĵĺ

-0.64

CONCLUS

-0.63

 Emin

-0.63

 scout

-0.62

 Totem

-0.62

 enthus

-0.62

POSITIVE LOGITS

amacare

1.02

 enshr

1.02

 altogether

1.00

 unconstitutional

0.97

 unilaterally

0.94

 expires

0.94

 statewide

0.93

 expire

0.93

 loopholes

0.92

 indefinitely

0.87

Activations Density 0.521%

words related to legal and policy changes

references to laws, regulations, and legal changes

No Comments

No Known Activations