INDEX

Explanations

phrases commonly found in email disclaimers

oai_token-act-pair · gpt-3.5-turbo

keywords associated with receiving news updates and promotional content via email

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

Gamer

-0.66

 Gamer

-0.64

 appell

-0.64

 Lilith

-0.63

hen

-0.63

 Painter

-0.61

gamer

-0.61

pai

-0.59

 loophole

-0.58

IGN

-0.57

POSITIVE LOGITS

interstitial

0.83

artifacts

0.74

ertility

0.74

currency

0.72

 transcripts

0.69

abytes

0.65

 Thumbnails

0.65

essential

0.64

 subscriptions

0.62

 debt

0.62

Activations Density 0.038%

phrases commonly found in email disclaimers

keywords associated with receiving news updates and promotional content via email

No Comments

No Known Activations