INDEX

Explanations

mentions of physicians in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to healthcare professionals, specifically physicians

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

flag

-0.85

eworld

-0.78

urat

-0.73

san

-0.72

erry

-0.71

runners

-0.71

lass

-0.70

forth

-0.69

dump

-0.69

 Orwell

-0.68

POSITIVE LOGITS

 physician

1.21

 physicians

1.08

 practitioner

1.01

 traged

0.92

 doctor

0.90

 Canaver

0.86

asus

0.86

 psychiatrist

0.86

 nursing

0.85

 care

0.85

Activations Density 0.011%

mentions of physicians in various contexts

references to healthcare professionals, specifically physicians

No Comments

No Known Activations