INDEX

Explanations

descriptions of events happening in the past

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

 intangible

-0.75

aum

-0.72

inventoryQuantity

-0.67

 symmetry

-0.66

éĹĺ

-0.65

afety

-0.63

 Gleaming

-0.62

MAC

-0.61

 hindsight

-0.61

TF

-0.61

POSITIVE LOGITS

ixties

0.85

aska

0.80

roth

0.77

ties

0.74

imester

0.73

xual

0.70

oice

0.68

nery

0.68

hours

0.66

assian

0.65

Activations Density 14.327%

descriptions of events happening in the past

No Comments

No Known Activations