INDEX

Explanations

mentions of the word "multiple."

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

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_slefr-ajt/2-res_slefr-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Reply

-0.95

umes

-0.87

Prince

-0.84

soDeliveryDate

-0.84

Mp

-0.82

sbm

-0.82

Â¨

-0.82

UME

-0.81

Benz

-0.80

hers

-0.80

POSITIVE LOGITS

xes

1.20

 sclerosis

1.13

 simultaneous

0.98

 multiple

0.96

 overlapping

0.87

 streams

0.86

 concurrent

0.85

 timelines

0.83

inguished

0.79

cro

0.79

Activations Density 0.065%

mentions of the word "multiple."

No Comments

No Known Activations