INDEX

Explanations

cryptocurrency-related terms

oai_token-act-pair · gpt-3.5-turbo

terms related to cryptocurrencies

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

ABE

-0.82

 Debor

-0.72

 Baxter

-0.69

ãĤī

-0.69

 Ange

-0.68

NEY

-0.68

 Faul

-0.67

Dek

-0.67

Alm

-0.66

 Borders

-0.66

POSITIVE LOGITS

coin

1.39

coins

1.26

rencies

1.09

Coin

1.01

folios

0.98

currency

0.96

bitcoin

0.95

apult

0.95

wallet

0.93

mint

0.91

Activations Density 0.005%

cryptocurrency-related terms

terms related to cryptocurrencies

No Comments

No Known Activations