INDEX

Explanations

medical terms or references

oai_token-act-pair · gpt-3.5-turbo

references to medical topics or conditions

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

xual

-0.99

uble

-0.87

unin

-0.81

cair

-0.74

yip

-0.73

Publisher

-0.71

wm

-0.71

plin

-0.71

arget

-0.71

ween

-0.69

POSITIVE LOGITS

 marijuana

1.11

 practition

1.07

 Marijuana

1.01

 examiner

0.96

 diagnosis

0.95

 care

0.95

 practitioner

0.93

 diagnoses

0.92

 doctor

0.92

 practitioners

0.89

Activations Density 0.037%

medical terms or references

references to medical topics or conditions

No Comments

No Known Activations