INDEX

Explanations

references to medical procedures

oai_token-act-pair · gpt-3.5-turbo

mentions of surgical procedures

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

anza

-0.79

spring

-0.77

izen

-0.74

 Prosper

-0.71

 Energy

-0.67

resp

-0.66

ership

-0.63

erd

-0.63

empty

-0.62

sheet

-0.62

POSITIVE LOGITS

 surgery

1.36

 Surgery

1.28

 surgeries

1.21

 surgeons

0.99

 amput

0.92

 surg

0.89

 medically

0.85

 anesthesia

0.84

 surgeon

0.84

 childbirth

0.83

Activations Density 0.017%

references to medical procedures

mentions of surgical procedures

No Comments

No Known Activations