INDEX

Explanations

references to box office performance and movie revenue

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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

uating

-1.16

uated

-1.14

uates

-1.14

ufact

-1.11

ittee

-1.08

vironment

-1.05

puter

-1.04

uate

-0.96

 Hots

-0.95

iva

-0.94

POSITIVE LOGITS

box

1.28

ridges

1.27

boxes

1.25

ers

1.25

cars

1.25

Box

1.25

buster

1.24

Box

1.20

 boxes

1.20

door

1.19

Activations Density 0.903%

references to box office performance and movie revenue

No Comments

No Known Activations