INDEX

Explanations

phrases related to medical conditions, especially those that are life-threatening

oai_token-act-pair · gpt-3.5-turbo

terms related to the severity of injuries, specifically those that are life-threatening

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

omet

-0.73

RA

-0.70

azon

-0.67

 Blueprint

-0.65

arist

-0.64

ometric

-0.64

spring

-0.64

 Revenue

-0.59

 bases

-0.59

rahim

-0.59

POSITIVE LOGITS

threatening

1.05

wcs

0.88

nces

0.79

elsius

0.76

xual

0.75

 aver

0.75

theless

0.75

terday

0.73

ciating

0.72

 diabetic

0.69

Activations Density 0.012%

phrases related to medical conditions, especially those that are life-threatening

terms related to the severity of injuries, specifically those that are life-threatening

No Comments

No Known Activations