INDEX

Explanations

mentions of nurses and related actions and attributes

oai_token-act-pair · gpt-3.5-turbo

references to nursing professionals or the term "nurse."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

çīĪ

-0.86

inations

-0.78

urat

-0.73

ãĥĪ

-0.73

RFC

-0.71

Ùħ

-0.71

izons

-0.70

Ø§ÙĦ

-0.68

atra

-0.68

arta

-0.67

POSITIVE LOGITS

maid

1.33

 practitioner

1.33

 nurse

1.09

 examiner

1.03

 nurs

0.99

 nurses

0.96

 Nurs

0.93

 practitioners

0.91

 nursing

0.87

 Nurse

0.79

Activations Density 0.010%

mentions of nurses and related actions and attributes

references to nursing professionals or the term "nurse."

No Comments

No Known Activations