INDEX

Explanations

names of people, places, and organizations

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

ILY

-0.60

ilogy

-0.57

ħĭ

-0.56

atform

-0.55

 transitional

-0.55

ancial

-0.54

ourced

-0.54

 symm

-0.53

 paradise

-0.52

 irony

-0.52

POSITIVE LOGITS

Ã±o

0.93

Ã±a

0.80

0.79

cker

0.77

witz

0.75

efe

0.74

llan

0.72

baugh

0.71

eh

0.70

ei

0.70

Activations Density 8.098%

names of people, places, and organizations

No Comments

No Known Activations