INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

BIP

-0.87

Newsletter

-0.81

lez

-0.77

Shape

-0.77

hower

-0.76

xtap

-0.73

notations

-0.71

schild

-0.71

interstitial

-0.68

Econom

-0.67

POSITIVE LOGITS

 fixme

0.71

 gland

0.71

 oxide

0.71

 cylinder

0.67

 cylinders

0.65

orage

0.65

 muzzle

0.64

 override

0.62

 glands

0.62

ocard

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.