INDEX

Explanations

references to medical patients and healthcare services

oai_token-act-pair · gpt-3.5-turbo

references to patients and patient-related terms in discussions about healthcare topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

angular

-0.86

ģ«

-0.82

auga

-0.80

oras

-0.80

dor

-0.76

eele

-0.75

redits

-0.75

argon

-0.74

gradient

-0.73

ordan

-0.73

POSITIVE LOGITS

 confidentiality

0.86

 patients

0.84

 patient

0.82

 Patient

0.82

 psychiatric

0.74

 illnesses

0.73

 Patients

0.73

killer

0.72

 care

0.71

ients

0.71

Activations Density 0.016%

references to medical patients and healthcare services

references to patients and patient-related terms in discussions about healthcare topics

No Comments

No Known Activations