INDEX

Explanations

mentions of mathematical equations or statements

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

 Gutenberg

-0.71

 Pixie

-0.69

 WARN

-0.66

 Wester

-0.65

 Universities

-0.65

AU

-0.65

 Primal

-0.64

ISM

-0.64

 OECD

-0.64

 Breach

-0.63

POSITIVE LOGITS

mur

1.24

geon

1.10

nit

1.03

phys

1.02

anda

1.01

abit

1.01

rats

0.99

asaki

0.98

daq

0.96

acles

0.94

Activations Density 6.327%

mentions of mathematical equations or statements

No Comments

No Known Activations