INDEX

Explanations

references to medical professionals, especially doctors

oai_token-act-pair · gpt-3.5-turbo

references to the profession of doctor

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

yss

-0.71

ndra

-0.71

 Huff

-0.71

 actionGroup

-0.70

Topics

-0.68

AW

-0.67

 Territories

-0.66

 Chamber

-0.62

 Hurricanes

-0.61

 Horn

-0.60

POSITIVE LOGITS

doctor

1.17

 doctor

1.16

 physician

0.94

 doctors

0.88

 practitioner

0.83

killer

0.82

abase

0.81

 examiner

0.81

 psychiatrist

0.79

 diagnoses

0.79

Activations Density 0.012%

references to medical professionals, especially doctors

references to the profession of doctor

No Comments

No Known Activations