INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Walter."

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

ebook

-0.99

arian

-0.95

ateur

-0.91

atories

-0.89

arians

-0.87

atory

-0.86

unity

-0.84

ional

-0.84

onal

-0.80

oral

-0.79

POSITIVE LOGITS

 Mond

0.93

 Reed

0.85

 Walter

0.84

 Winc

0.79

 Cron

0.76

 Payton

0.76

\\\\\\\\

0.75

 Isaac

0.74

Ley

0.73

Sob

0.73

Activations Density 0.012%

proper nouns

mentions of the name "Walter."

No Comments

No Known Activations