INDEX

Explanations

titles or sequels in series

oai_token-act-pair · gpt-3.5-turbo

references to sequels or series of works

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

onday

-0.85

doors

-0.75

gov

-0.70

anced

-0.68

fax

-0.67

vana

-0.65

anca

-0.65

yip

-0.64

 prevailed

-0.63

 corners

-0.63

POSITIVE LOGITS

 series

1.37

 trilogy

1.34

series

1.22

 saga

1.14

 Series

1.06

Series

1.04

 Trilogy

0.93

 franchise

0.86

 Cycle

0.84

 illustrious

0.84

Activations Density 0.187%

titles or sequels in series

references to sequels or series of works

No Comments

No Known Activations