INDEX

Explanations

healthcare-related terms, specifically related to legislative acts and care organizations

oai_token-act-pair · gpt-3.5-turbo

references to healthcare legislation and related policies

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

nav

-0.64

 hiding

-0.64

 perjury

-0.64

 Kenyan

-0.63

kered

-0.61

ãĥĥãĥī

-0.61

SN

-0.60

KEN

-0.60

integer

-0.59

toggle

-0.59

POSITIVE LOGITS

taker

1.19

 Care

1.18

care

1.12

Care

0.96

 Provider

0.89

 Clinic

0.86

 Aware

0.84

 Patient

0.82

payer

0.81

croft

0.80

Activations Density 0.009%

healthcare-related terms, specifically related to legislative acts and care organizations

references to healthcare legislation and related policies

No Comments

No Known Activations