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

 Tourism

-0.65

 coastline

-0.64

 Tigers

-0.63

 linem

-0.62

 Rutgers

-0.62

OD

-0.61

 targ

-0.60

 Ferry

-0.59

 Exile

-0.59

 Jord

-0.59

POSITIVE LOGITS

´

0.81

pees

0.79

Ĥİ

0.79

elight

0.78

¸

0.78

²¾

0.78

ãĤ°

0.78

¶

0.71

ãĥīãĥ©ãĤ´ãĥ³

0.70

Page

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.