INDEX

Explanations

instances of the word "Kel" or "Kol" in the text

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names, related to individuals and organizations

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

 AMERICA

-0.68

CLASSIFIED

-0.68

xual

-0.66

 Sakuya

-0.65

enegger

-0.65

ItemTracker

-0.65

 Afric

-0.64

displayText

-0.64

 accur

-0.64

reads

-0.63

POSITIVE LOGITS

tering

1.09

seys

1.03

sten

0.98

tered

0.95

inda

0.92

sey

0.92

zeb

0.91

istani

0.90

ter

0.90

pie

0.89

Activations Density 0.009%

instances of the word "Kel" or "Kol" in the text

proper nouns, particularly names, related to individuals and organizations

No Comments

No Known Activations