INDEX

Explanations

words related to specific locations or organizations, particularly those containing "Rac."

oai_token-act-pair · gpt-3.5-turbo

mentions of "Racine" and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ITNESS

-0.78

 Kush

-0.75

 Petraeus

-0.75

 ç¥ŀ

-0.74

è£ħ

-0.70

utterstock

-0.69

amsung

-0.67

HK

-0.64

 Hawaiian

-0.63

ãĢĲ

-0.62

POSITIVE LOGITS

quet

1.15

rac

1.05

rac

1.03

coon

1.02

oust

1.02

kees

0.99

eways

0.98

eway

0.96

ials

0.95

emate

0.93

Activations Density 0.008%

words related to specific locations or organizations, particularly those containing "Rac."

mentions of "Racine" and related terms

No Comments

No Known Activations