INDEX

Explanations

words related to healthcare, specifically nursing homes

oai_token-act-pair · gpt-3.5-turbo

references to nursing and caregiving environments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 attm

-0.76

inatory

-0.75

Ùħ

-0.74

ãĥĪ

-0.70

upon

-0.67

inations

-0.67

izoph

-0.67

 Ramadan

-0.66

xxxxxxxx

-0.66

SX

-0.65

POSITIVE LOGITS

 nursing

1.03

maid

0.92

 practitioner

0.87

otin

0.80

hurst

0.80

 practition

0.75

yard

0.74

 lact

0.73

 Nursing

0.72

 care

0.72

Activations Density 0.012%

words related to healthcare, specifically nursing homes

references to nursing and caregiving environments

No Comments

No Known Activations