INDEX

Explanations

phrases related to giving suggestions or advice

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

 Beck

-0.60

 ãĤµ

-0.55

 Prosecut

-0.50

ĸļ

-0.47

umerous

-0.47

 Investigative

-0.46

eatures

-0.45

roma

-0.45

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.45

netflix

-0.45

POSITIVE LOGITS

uate

0.64

 things

0.59

ezvous

0.58

rity

0.55

hello

0.52

 dealing

0.52

grass

0.51

 deal

0.50

itaire

0.50

awaru

0.49

Activations Density 12.935%

phrases related to giving suggestions or advice

No Comments

No Known Activations