INDEX

Explanations

locations related to medical emergencies

oai_token-act-pair · gpt-3.5-turbo

phrases related to hospitalizations and medical treatment following injuries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 diameter

-0.67

 preferring

-0.62

 Especially

-0.60

 gestures

-0.59

goal

-0.58

 Shutterstock

-0.58

coal

-0.57

gran

-0.57

 Tanz

-0.57

 Links

-0.57

POSITIVE LOGITS

 awaited

0.78

estyles

0.76

 await

0.74

 disinfect

0.73

 awaiting

0.72

 interrogation

0.72

upon

0.72

hett

0.71

 interrogated

0.70

axter

0.69

Activations Density 0.303%

locations related to medical emergencies

phrases related to hospitalizations and medical treatment following injuries

No Comments

No Known Activations