INDEX

Explanations

dates or years mentioned in the text

oai_token-act-pair · gpt-3.5-turbo

the presence of empty tokens or sections with no content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 disadvant

-0.70

oÄŁ

-0.65

iste

-0.61

 nearby

-0.58

 kios

-0.57

 voic

-0.56

amaru

-0.56

 fingert

-0.56

Alternatively

-0.55

dial

-0.55

POSITIVE LOGITS

Reviewer

0.77

BuyableInstoreAndOnline

0.69

 Released

0.65

released

0.64

DragonMagazine

0.63

 revised

0.61

 Recession

0.61

bernatorial

0.60

 Carnegie

0.57

 Grateful

0.57

Activations Density 0.650%

dates or years mentioned in the text

the presence of empty tokens or sections with no content

No Comments

No Known Activations