INDEX

Explanations

mentions of the name "De" followed by other characters

oai_token-act-pair · gpt-3.5-turbo

references to the name "De."

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

ĸļ

-0.97

uyomi

-0.84

ãĤ¤ãĥĪ

-0.80

DragonMagazine

-0.77

 sidx

-0.77

INGS

-0.73

hetti

-0.72

CPC

-0.69

ãĤ¼ãĤ¦ãĤ¹

-0.68

allery

-0.67

POSITIVE LOGITS

legate

1.17

leted

1.12

utsch

1.09

legates

1.09

arest

1.08

vers

1.05

classified

1.03

bris

1.02

uter

0.96

VICE

0.95

Activations Density 0.018%

mentions of the name "De" followed by other characters

references to the name "De."

No Comments

No Known Activations