INDEX

Explanations

references to readers or audience of content

oai_token-act-pair · gpt-3.5-turbo

references to readers and their engagement

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Pharma

-0.67

 Skull

-0.66

 Bain

-0.65

Alb

-0.65

Spa

-0.64

asia

-0.64

 Sabb

-0.64

ering

-0.61

 Romo

-0.61

 Guan

-0.60

POSITIVE LOGITS

hip

1.52

hips

0.93

uggest

0.89

boy

0.87

DragonMagazine

0.86

boys

0.85

 reader

0.83

0.82

pace

0.81

iences

0.79

Activations Density 0.027%

references to readers or audience of content

references to readers and their engagement

No Comments

No Known Activations