INDEX

Explanations

information or locations related to central hubs

oai_token-act-pair · gpt-3.5-turbo

references to centers of activity or organization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

olin

-0.78

ãģĭ

-0.74

idate

-0.70

Integer

-0.68

ISH

-0.67

ormal

-0.66

anda

-0.66

oln

-0.66

 exorc

-0.66

restrial

-0.64

POSITIVE LOGITS

hub

3.91

 hubs

2.96

Hub

2.36

hub

2.17

Hub

1.72

 nexus

1.18

 gateway

1.10

cog

1.03

 hots

1.00

 hive

0.99

Activations Density 0.017%

information or locations related to central hubs

references to centers of activity or organization

No Comments

No Known Activations