INDEX

Explanations

names of people and places, particularly in the context of news articles

oai_token-act-pair · gpt-3.5-turbo

textual elements that indicate context or commentary related to significant events or decisions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Sloven

-0.75

 Hispan

-0.73

 Homs

-0.73

 Budapest

-0.71

 Slovenia

-0.67

 Yanukovych

-0.67

chwitz

-0.67

 Prague

-0.66

 Croatian

-0.66

 Czech

-0.65

POSITIVE LOGITS

Austral

1.08

Australian

1.04

 Adelaide

1.02

 Australians

1.02

Australia

1.01

elaide

0.97

 Queensland

0.95

 Canberra

0.91

 Perth

0.91

 Tasman

0.90

Activations Density 0.128%

names of people and places, particularly in the context of news articles

textual elements that indicate context or commentary related to significant events or decisions

No Comments

No Known Activations