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

bnb

-0.86

NetMessage

-0.82

unda

-0.71

stem

-0.68

ILA

-0.67

conservancy

-0.66

rim

-0.65

utra

-0.64

elsius

-0.63

 relax

-0.63

POSITIVE LOGITS

udeb

0.72

 compos

0.69

ath

0.65

 sketches

0.65

 Factor

0.65

hor

0.64

Tir

0.63

ol

0.62

inous

0.61

Hor

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.