INDEX

Explanations

phrases related to exaggeration and overstatement

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

arov

-0.49

rollers

-0.48

ovych

-0.47

 Aether

-0.46

 erect

-0.46

olen

-0.46

Ķ

-0.46

Hep

-0.45

abiding

-0.45

forums

-0.45

POSITIVE LOGITS

 understatement

0.72

ously

0.64

 guiActiveUn

0.63

uphem

0.59

 msec

0.57

isSpecialOrderable

0.57

 exaggeration

0.56

Osw

0.55

atures

0.55

 estim

0.55

Activations Density 15.833%

phrases related to exaggeration and overstatement

No Comments

No Known Activations