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

hester

-0.76

ratulations

-0.76

oeuv

-0.72

 horizont

-0.71

izont

-0.69

 actionGroup

-0.68

ripp

-0.68

yrinth

-0.68

ritical

-0.67

Ô

-0.67

POSITIVE LOGITS

 Summers

0.71

Mae

0.69

 Healer

0.66

onement

0.65

Oracle

0.65

Synopsis

0.64

 Tire

0.63

Script

0.62

 Celeb

0.62

ypes

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.