INDEX

Explanations

buzzwords or phrases related to recent or current events or topics

oai_token-act-pair · gpt-3.5-turbo

mentions of new developments or reports

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

 actionGroup

-0.84

cius

-0.81

ashtra

-0.79

ufact

-0.76

xual

-0.75

Zip

-0.69

cientious

-0.69

oleon

-0.68

Reloaded

-0.67

omever

-0.66

POSITIVE LOGITS

bie

1.24

bies

1.19

 arrivals

0.99

 developments

0.93

 additions

0.91

 revelations

0.89

 Zealand

0.88

foundland

0.87

 revelation

0.86

 YORK

0.85

Activations Density 0.098%

buzzwords or phrases related to recent or current events or topics

mentions of new developments or reports

No Comments

No Known Activations