INDEX

Explanations

medical professionals, specifically doctors

oai_token-act-pair · gpt-3.5-turbo

references to doctors and medical professionals

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

 Territories

-0.95

 Flavoring

-0.66

ptions

-0.63

 Dwar

-0.63

wm

-0.63

doors

-0.63

 Hornets

-0.60

 Strongh

-0.58

 Tactics

-0.57

plings

-0.57

POSITIVE LOGITS

 prescribing

1.02

 specializing

0.88

 treating

0.87

 prescribe

0.87

 doctor

0.84

 examiner

0.81

 diagn

0.81

 diagnosed

0.79

icians

0.79

doctor

0.79

Activations Density 0.042%

medical professionals, specifically doctors

references to doctors and medical professionals

No Comments

No Known Activations