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

 Anthem

-0.69

 Minutes

-0.62

 herald

-0.61

 antagonist

-0.60

 leveled

-0.60

 Challenge

-0.58

 Versus

-0.57

 Cooperative

-0.57

 NAME

-0.56

 antagonists

-0.55

POSITIVE LOGITS

umerable

0.87

igraph

0.82

 calendars

0.79

bidden

0.78

orem

0.75

agos

0.73

auer

0.72

izen

0.71

ogg

0.70

paio

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.