INDEX

Explanations

No Explanations Found

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

Lord

-0.73

 Cutter

-0.69

 Chandra

-0.68

=-=-=-=-=-=-=-=-

-0.68

Magikarp

-0.67

 Aram

-0.67

axe

-0.66

Stone

-0.65

 Throne

-0.62

stack

-0.62

POSITIVE LOGITS

itures

0.76

 electr

0.64

acons

0.62

 pairs

0.62

mot

0.61

 clubs

0.61

 bets

0.61

zbek

0.60

uese

0.60

 Clubs

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.