INDEX

Explanations

phrases related to healthcare situations, particularly relating to nursing and nursing homes

oai_token-act-pair · gpt-3.5-turbo

references to nurses and nursing professions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

inations

-0.86

 Ramadan

-0.78

izons

-0.77

inates

-0.75

çīĪ

-0.74

azeera

-0.67

ogh

-0.66

Ùħ

-0.65

urat

-0.65

itious

-0.64

POSITIVE LOGITS

maid

1.50

 practitioner

1.32

 nurs

1.00

 nurse

0.94

 nursing

0.91

 practitioners

0.88

 examiner

0.81

 Nurs

0.81

 nurses

0.79

hop

0.79

Activations Density 0.031%

phrases related to healthcare situations, particularly relating to nursing and nursing homes

references to nurses and nursing professions

No Comments

No Known Activations