INDEX

Explanations

TV show titles and broadcast times

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

etheless

-0.60

ificantly

-0.56

ogyn

-0.55

 intrins

-0.55

schild

-0.55

 Sidd

-0.55

omorphic

-0.54

swers

-0.54

odox

-0.53

gha

-0.53

POSITIVE LOGITS

0.88

0.70

0.64

Pocket

0.64

0.63

 Poster

0.62

 Song

0.61

 Printed

0.61

TBD

0.61

0.60

Activations Density 8.192%

TV show titles and broadcast times

No Comments

No Known Activations