INDEX

Explanations

keywords related to images or visuals like "slide" or "PowerPoint"

oai_token-act-pair · gpt-3.5-turbo

references to slides in a presentation or display

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

orate

-0.79

 Carbuncle

-0.75

omes

-0.75

etheless

-0.73

raising

-0.70

elf

-0.70

ãĥ¯ãĥ³

-0.67

ilee

-0.66

omers

-0.65

rises

-0.64

POSITIVE LOGITS

 slide

0.85

itzer

0.82

 Canaver

0.81

boards

0.81

 slides

0.80

board

0.77

 slope

0.75

creen

0.74

ulic

0.72

uten

0.72

Activations Density 0.028%

keywords related to images or visuals like "slide" or "PowerPoint"

references to slides in a presentation or display

No Comments

No Known Activations