INDEX

Explanations

terms related to marketing

oai_token-act-pair · gpt-3.5-turbo

terms related to marketing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

-0.84

inite

-0.70

Fas

-0.69

 Templar

-0.65

 captcha

-0.65

 Carbuncle

-0.65

upon

-0.65

 TAMADRA

-0.65

 Fulton

-0.63

tar

-0.62

POSITIVE LOGITS

eering

0.96

 campaigns

0.93

 ploy

0.86

eers

0.83

 marketing

0.82

 gimm

0.81

 gimmick

0.80

 strategy

0.80

 indo

0.79

 prowess

0.78

Activations Density 0.023%

terms related to marketing

terms related to marketing

No Comments

No Known Activations