INDEX

Explanations

proper nouns related to locations or events

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Washington."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ãĥł

-0.73

oat

-0.63

antha

-0.60

ãĥ¼ãĥĨãĤ£

-0.59

formation

-0.59

 append

-0.56

cible

-0.56

inges

-0.56

 nearest

-0.55

ultan

-0.55

POSITIVE LOGITS

 CITY

1.17

WASHINGTON

1.03

ASHINGTON

1.02

 COUNTY

0.96

 WEEK

0.95

 âĢķ

0.94

 GOODMAN

0.92

TIM

0.91

MAN

0.90

 IMAGES

0.90

Activations Density 0.029%

proper nouns related to locations or events

occurrences of the word "Washington."

No Comments

No Known Activations