INDEX

Explanations

The main thing this neuron does is find mentions of a specific location, namely "Flint"

oai_token-act-pair · gpt-3.5-turbo

mentions of the city of Flint, especially in relation to the water crisis

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

-0.90

mble

-0.74

asin

-0.73

Magikarp

-0.71

ISM

-0.71

VIDIA

-0.69

*/(

-0.68

 satell

-0.67

uner

-0.67

 magnification

-0.66

POSITIVE LOGITS

stones

1.27

shire

1.03

stone

0.91

 Flint

0.90

lock

0.81

lings

0.78

ridge

0.78

brook

0.77

wall

0.74

street

0.72

Activations Density 0.011%

The main thing this neuron does is find mentions of a specific location, namely "Flint"

mentions of the city of Flint, especially in relation to the water crisis

No Comments

No Known Activations