INDEX

Explanations

phrases related to detachment or disconnection

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

gio

-0.58

GB

-0.56

eta

-0.55

xual

-0.55

 intimid

-0.54

iola

-0.54

 graz

-0.54

Ô

-0.54

 abund

-0.52

esan

-0.52

POSITIVE LOGITS

 disconnect

0.72

yip

0.68

owship

0.66

omorph

0.64

ribut

0.64

alin

0.63

aline

0.61

 detachment

0.60

Mub

0.59

 Schr

0.59

Activations Density 5.452%

phrases related to detachment or disconnection

No Comments

No Known Activations