INDEX

Explanations

medical conditions, legal terms, and terms related to community

oai_token-act-pair · gpt-3.5-turbo

terms related to legal and medical concepts

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

psons

-0.56

ulhu

-0.56

*/(

-0.54

ften

-0.53

ovie

-0.53

agonist

-0.53

rower

-0.53

ndum

-0.52

agame

-0.51

akeru

-0.51

POSITIVE LOGITS

bugs

0.53

 purposes

0.52

 margins

0.52

 warfare

0.51

 emergencies

0.50

 levels

0.50

 proportions

0.50

 traders

0.48

 considerations

0.48

 agriculture

0.47

Activations Density 1.188%

medical conditions, legal terms, and terms related to community

terms related to legal and medical concepts

No Comments

No Known Activations