INDEX

Explanations

proper nouns related to geographical locations

oai_token-act-pair · gpt-3.5-turbo

mentions of the location Durham

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Holo

-0.91

Laz

-0.84

rub

-0.83

Mes

-0.81

 healer

-0.79

Mak

-0.78

 Magicka

-0.77

 armour

-0.76

hero

-0.75

mage

-0.75

POSITIVE LOGITS

 Durham

2.61

 Raleigh

2.40

 Cornell

2.04

 Asheville

1.82

 Chapel

1.68

 Cary

1.66

 Wilmington

1.63

 Triangle

1.62

 Burr

1.60

UNC

1.53

Activations Density 0.051%

proper nouns related to geographical locations

mentions of the location Durham

No Comments

No Known Activations