INDEX

Explanations

texts pertaining to detailed information about events, processes, or logistics

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

anni

-0.70

lish

-0.69

rones

-0.69

Mar

-0.67

jer

-0.66

Favorite

-0.65

Medium

-0.65

MAR

-0.64

soever

-0.64

iel

-0.63

POSITIVE LOGITS

 circumstances

1.07

 context

0.89

 circumstance

0.83

 precedent

0.81

 actionGroup

0.80

 proximity

0.79

nces

0.78

 constraints

0.78

 propensity

0.77

 precedence

0.75

Activations Density 8.924%

texts pertaining to detailed information about events, processes, or logistics

No Comments

No Known Activations