INDEX

Explanations

mentions of medical professionals, particularly doctors

oai_token-act-pair · gpt-3.5-turbo

mentions of doctors

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

 Territories

-0.81

 Flavoring

-0.76

iland

-0.64

cluding

-0.63

CHAT

-0.63

âĸ¬

-0.62

yss

-0.61

restricted

-0.60

Characters

-0.60

ndra

-0.59

POSITIVE LOGITS

doctor

1.03

 doctor

1.01

 examiner

0.96

 practitioner

0.86

 physician

0.83

ysis

0.82

 Doctor

0.82

 diagn

0.80

 diagnoses

0.80

 prescribing

0.78

Activations Density 0.024%

mentions of medical professionals, particularly doctors

mentions of doctors

No Comments

No Known Activations