INDEX

Explanations

names or terms related to businesses or organizations

oai_token-act-pair · gpt-3.5-turbo

the letter "D" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Reloaded

-0.68

ften

-0.67

thood

-0.66

ãĤª

-0.63

 streng

-0.63

lime

-0.62

ItemThumbnailImage

-0.61

FontSize

-0.61

eal

-0.60

anyahu

-0.60

POSITIVE LOGITS

2.91

1.90

1.57

DV

1.26

DC

1.24

DA

1.20

1.19

DG

1.16

DD

1.15

Di

1.12

Activations Density 0.038%

names or terms related to businesses or organizations

the letter "D" in various contexts

No Comments

No Known Activations