INDEX

Explanations

neighborhood-related terms or locations

oai_token-act-pair · gpt-3.5-turbo

mentions of neighborhoods or related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 bearer

-0.80

REDACTED

-0.72

udeb

-0.70

 Mehran

-0.68

ATA

-0.67

ISM

-0.65

utenberg

-0.64

))))

-0.63

xxxxxxxx

-0.60

ista

-0.60

POSITIVE LOGITS

bors

1.44

bour

1.12

bor

1.11

Neigh

0.95

 Neigh

0.95

eties

0.91

stairs

0.91

«

0.90

 neighb

0.89

 neighbour

0.88

Activations Density 0.007%

neighborhood-related terms or locations

mentions of neighborhoods or related terms

No Comments

No Known Activations