INDEX

Explanations

mentions related to medical professionals such as doctors, clinicians, and professors in medical fields

oai_token-act-pair · gpt-3.5-turbo

professional titles and roles in the medical field

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

notations

-0.76

ebin

-0.74

alion

-0.73

ascript

-0.73

estamp

-0.72

itiveness

-0.71

pload

-0.70

ffiti

-0.69

eanor

-0.67

tarians

-0.67

POSITIVE LOGITS

 pediatric

1.63

 Pediatrics

1.54

 neurolog

1.53

 Neurolog

1.51

 Clinical

1.50

 Surgery

1.50

 psychiatry

1.49

 Diabetes

1.48

 Physicians

1.48

 gastro

1.48

Activations Density 0.298%

mentions related to medical professionals such as doctors, clinicians, and professors in medical fields

professional titles and roles in the medical field

No Comments

No Known Activations