INDEX

Explanations

the prefix "ar" followed by varying suffixes

oai_token-act-pair · gpt-3.5-turbo

references to regions or areas, particularly those starting with "ar"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

å§«

-0.80

éĹĺ

-0.77

 Seller

-0.70

HAEL

-0.68

assetsadobe

-0.67

¬¼

-0.67

ĸļ

-0.65

aneers

-0.65

ega

-0.63

 Prosper

-0.62

POSITIVE LOGITS

beit

1.19

thur

1.10

ansas

1.02

duino

0.97

rog

0.97

ctic

0.97

atars

0.90

rival

0.90

agon

0.88

chery

0.88

Activations Density 0.011%

the prefix "ar" followed by varying suffixes

references to regions or areas, particularly those starting with "ar"

No Comments

No Known Activations