INDEX

Explanations

references to academic articles and their authors

oai_token-act-pair · gpt-3.5-turbo

references to academic citations and authorship

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 comprom

-0.73

FactoryReloaded

-0.70

 Username

-0.70

 Doctrine

-0.69

ropolitan

-0.68

 Jobs

-0.68

Ner

-0.63

 Maker

-0.63

 Corps

-0.63

 Shades

-0.62

POSITIVE LOGITS

.,

1.08

icably

0.87

manac

0.87

phys

0.86

ittle

0.84

ibi

0.81

herty

0.79

abo

0.78

ibaba

0.77

leys

0.77

Activations Density 0.018%

references to academic articles and their authors

references to academic citations and authorship

No Comments

No Known Activations