INDEX

Explanations

specific words related to computer software or features

oai_token-act-pair · gpt-3.5-turbo

instances of the word "biv"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

--------------------------------------------------------

-0.84

 secrecy

-0.63

STER

-0.63

QU

-0.62

 unregulated

-0.61

cffffcc

-0.61

 tradem

-0.60

 overpower

-0.60

————

-0.60

xon

-0.60

POSITIVE LOGITS

iew

1.13

isions

1.04

irus

1.03

isible

1.01

iral

1.00

iv

0.97

iera

0.96

olution

0.96

entric

0.94

otal

0.94

Activations Density 0.013%

specific words related to computer software or features

instances of the word "biv"

No Comments

No Known Activations