INDEX

Explanations

mentions of legal terms or concepts

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "law."

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

izoph

-0.94

 Scion

-0.72

 Flavoring

-0.68

 periphery

-0.68

Dak

-0.67

rican

-0.66

vae

-0.64

 unarmed

-0.64

 Afric

-0.63

NFC

-0.63

POSITIVE LOGITS

yers

1.20

law

1.18

laws

1.17

suits

1.11

yer

1.07

suit

1.06

eering

0.99

makers

0.97

iard

0.96

ski

0.94

Activations Density 0.005%

mentions of legal terms or concepts

occurrences of the word "law."

No Comments

No Known Activations