INDEX

Explanations

phrases containing information about a person's professional history

oai_token-act-pair · gpt-3.5-turbo

phrases indicating prior experiences or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

umeric

-0.83

pmwiki

-0.83

=-=-=-=-=-=-=-=-

-0.80

aptic

-0.76

Detailed

-0.71

vous

-0.68

ivably

-0.66

umer

-0.65

abus

-0.64

Supp

-0.63

POSITIVE LOGITS

 joining

1.33

 retiring

1.26

 quitting

1.21

 stint

1.20

 graduating

1.20

 retirement

1.14

 resign

1.12

 founding

1.11

 becoming

1.10

 marrying

1.02

Activations Density 0.209%

phrases containing information about a person's professional history

phrases indicating prior experiences or events

No Comments

No Known Activations