INDEX

Explanations

code implementations and programming language concepts related to implementing classes, methods, interfaces, and functions

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

fare

-0.45

uyomi

-0.43

deb

-0.40

watch

-0.36

claim

-0.35

onna

-0.33

birth

-0.33

 wiped

-0.33

few

-0.32

wagon

-0.32

POSITIVE LOGITS

urally

0.38

ATIONAL

0.37

ations

0.37

ITNESS

0.37

 Prototype

0.37

emonium

0.36

 Architect

0.36

itives

0.36

ments

0.35

ORE

0.35

Activations Density 10.405%

code implementations and programming language concepts related to implementing classes, methods, interfaces, and functions

No Comments

No Known Activations