INDEX

Explanations

The main thing this neuron does is find names or words related to particular individuals

oai_token-act-pair · gpt-3.5-turbo

specific individual names and related identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Nichols

-0.86

 Nich

-0.84

Nich

-0.81

 Molly

-0.79

 Mania

-0.77

 Pilgrim

-0.76

 Wass

-0.76

 Nicotine

-0.75

 Nicholson

-0.74

Wid

-0.73

POSITIVE LOGITS

ar

1.80

AR

1.65

ars

1.45

aran

1.28

arp

1.27

AR

1.25

aro

1.25

aru

1.23

arat

1.21

arr

1.20

Activations Density 0.294%

The main thing this neuron does is find names or words related to particular individuals

specific individual names and related identifiers

No Comments

No Known Activations