INDEX

Explanations

health-related terms related to medical conditions and bodily functions, specifically focusing on cancer risks and thyroid issues

oai_token-act-pair · gpt-3.5-turbo

terms related to medical conditions and cancers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Tradable

-1.02

ovie

-0.92

hig

-0.84

ombie

-0.82

aneers

-0.78

DragonMagazine

-0.75

Original

-0.75

ÅĤ

-0.75

 Bravo

-0.74

baugh

-0.72

POSITIVE LOGITS

 gland

1.22

 disorders

1.09

 asthma

1.08

 thyroid

1.07

muc

1.07

 symptoms

1.06

 hormone

1.05

 inflammation

1.04

 hormones

1.03

 hormonal

1.02

Activations Density 0.097%

health-related terms related to medical conditions and bodily functions, specifically focusing on cancer risks and thyroid issues

terms related to medical conditions and cancers

No Comments

No Known Activations