INDEX

Explanations

technical descriptions of prototypes and testing programs

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

olulu

-0.76

arching

-0.69

andra

-0.69

aiman

-0.66

wind

-0.64

icago

-0.64

hani

-0.62

vation

-0.62

sole

-0.61

bourne

-0.60

POSITIVE LOGITS

 prototype

0.72

 prototypes

0.69

hesis

0.65

 Prototype

0.63

smanship

0.62

 prototyp

0.58

 template

0.58

 Mock

0.56

 sketches

0.55

 sketch

0.54

Activations Density 7.565%

technical descriptions of prototypes and testing programs

No Comments

No Known Activations