INDEX

Explanations

phrases related to making decisions or choices

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

Introduced

-0.68

mone

-0.62

via

-0.59

###

-0.58

shall

-0.57

April

-0.56

 vividly

-0.56

wordpress

-0.56

Â§

-0.56

angelo

-0.55

POSITIVE LOGITS

 hire

0.74

bidden

0.71

gotten

0.69

 precaution

0.66

geries

0.66

 example

0.65

raviolet

0.65

WARD

0.63

 captcha

0.62

 starters

0.61

Activations Density 12.975%

phrases related to making decisions or choices

No Comments

No Known Activations