INDEX

Explanations

stage directions and technical instructions in a written work

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

vironment

-0.83

eros

-0.82

umar

-0.82

vernment

-0.78

anguages

-0.77

tymology

-0.75

ipop

-0.73

zzle

-0.73

rontal

-0.71

 Bears

-0.71

POSITIVE LOGITS

craft

1.13

yard

0.96

Stage

0.90

room

0.86

roller

0.86

rooms

0.86

stage

0.84

\\\\

0.83

igmatic

0.80

building

0.80

Activations Density 6.870%

stage directions and technical instructions in a written work

No Comments

No Known Activations