INDEX

Explanations

locations mentioned in text

oai_token-act-pair · gpt-3.5-turbo

phrases indicating specific locations or addresses

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

icter

-0.70

 haircut

-0.70

casters

-0.68

 pals

-0.68

bush

-0.65

sense

-0.62

INESS

-0.61

vous

-0.60

 hairc

-0.59

IENCE

-0.59

POSITIVE LOGITS

 Burr

0.75

 Gothic

0.71

 Eucl

0.71

ropolitan

0.70

 Richmond

0.69

 Southeast

0.69

GMT

0.68

 northeastern

0.67

 Surrey

0.66

 geographically

0.66

Activations Density 0.233%

locations mentioned in text

phrases indicating specific locations or addresses

No Comments

No Known Activations