INDEX

Explanations

phrases or sentences that mention something intriguing or noteworthy

oai_token-act-pair · gpt-3.5-turbo

discussions expressing curiosity or intrigue about various topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

helle

-0.80

oise

-0.75

aper

-0.75

avers

-0.75

reditation

-0.72

ithing

-0.71

ussy

-0.71

uts

-0.69

redited

-0.68

rapnel

-0.67

POSITIVE LOGITS

tid

0.93

 Flavoring

0.86

arios

0.84

 twists

0.84

 sidel

0.80

 insights

0.74

ly

0.74

 trivia

0.74

 anecdotes

0.74

 developments

0.71

Activations Density 0.048%

phrases or sentences that mention something intriguing or noteworthy

discussions expressing curiosity or intrigue about various topics

No Comments

No Known Activations