INDEX

Explanations

No Explanations Found

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

Austral

-0.77

edin

-0.71

ãĥ´ãĤ¡

-0.71

anie

-0.69

 Cheong

-0.68

 Horus

-0.67

 largeDownload

-0.67

ogi

-0.67

 Sang

-0.66

heng

-0.66

POSITIVE LOGITS

âĵĺ

0.68

ffer

0.67

 antitrust

0.64

ishop

0.63

 omit

0.62

erella

0.60

 missionary

0.60

ordering

0.59

 theft

0.59

witch

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.