INDEX

Explanations

medical diagnoses

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of medical diagnoses and diagnostic processes

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

atility

-0.64

Leg

-0.63

yss

-0.63

asin

-0.62

sen

-0.62

hire

-0.62

 reserve

-0.61

 incorporation

-0.61

sol

-0.61

committee

-0.60

POSITIVE LOGITS

 Diagn

1.27

 diagnoses

1.25

ostics

1.21

 diagnosis

1.21

 diagn

1.06

diagn

1.01

 diagnosed

1.00

 diagnose

0.98

ptoms

0.90

ostic

0.89

Activations Density 0.007%

medical diagnoses

repeated mentions of medical diagnoses and diagnostic processes

No Comments

No Known Activations