INDEX

Explanations

proper nouns, particularly those related to political figures and events

oai_token-act-pair · gpt-3.5-turbo

specific political figures or titles associated with events

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

Philadelphia

-0.76

crew

-0.67

 chem

-0.65

Seattle

-0.65

lectic

-0.65

 Midwest

-0.64

 harmonic

-0.64

 webs

-0.64

CRE

-0.63

Denver

-0.63

POSITIVE LOGITS

jri

1.30

oglu

1.12

ÄŁ

1.12

oÄŁ

1.10

Sr

0.95

 presided

0.95

Vaj

0.94

 resigned

0.92

ovich

0.92

llah

0.90

Activations Density 0.238%

proper nouns, particularly those related to political figures and events

specific political figures or titles associated with events

No Comments

No Known Activations