INDEX

Explanations

references to the word "Gold"

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "Gold."

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

ATIONS

-0.90

SPONSORED

-0.78

ROR

-0.67

ect

-0.67

mediated

-0.64

actic

-0.63

ENDED

-0.62

 constitu

-0.62

uated

-0.60

ched

-0.60

POSITIVE LOGITS

 medal

1.01

gold

0.92

 coins

0.92

 medals

0.91

smith

0.90

mining

0.89

arton

0.88

wyn

0.88

estone

0.88

coin

0.88

Activations Density 0.005%

references to the word "Gold"

mentions of the word "Gold."

No Comments

No Known Activations