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

terday

-0.91

conn

-0.76

cham

-0.76

engeance

-0.75

pas

-0.74

Downloadha

-0.74

laws

-0.73

 forb

-0.69

hin

-0.67

 Winc

-0.66

POSITIVE LOGITS

 Center

1.49

Center

1.18

 center

1.16

 centers

0.99

center

0.93

 Centers

0.91

 Centre

0.81

 centre

0.78

 Provider

0.71

REF

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.