INDEX

Explanations

information related to prize offerings and numerical values like amounts and quantities

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

ħĭ

-0.75

alogue

-0.70

ESE

-0.70

ilda

-0.70

 Nieto

-0.69

 brut

-0.68

 Dickinson

-0.68

 kindly

-0.67

Exit

-0.67

 commanding

-0.67

POSITIVE LOGITS

pees

0.99

abytes

0.96

coins

0.94

Downloadha

0.93

 pounds

0.90

votes

0.86

MSN

0.83

lbs

0.81

acht

0.78

embed

0.75

Activations Density 19.763%

information related to prize offerings and numerical values like amounts and quantities

No Comments

No Known Activations