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

pez

-0.81

estern

-0.74

phant

-0.74

ierrez

-0.70

ugal

-0.67

 Pixie

-0.66

qa

-0.66

cyclopedia

-0.65

ugi

-0.64

aciously

-0.63

POSITIVE LOGITS

Preview

0.76

AppData

0.74

Privacy

0.73

hots

0.72

Installation

0.67

nod

0.64

DOWN

0.64

 Clapper

0.63

Status

0.63

ertodd

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.