INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Pupp

-0.76

 lett

-0.72

fing

-0.66

 mosqu

-0.65

 Pepe

-0.63

 Gram

-0.63

 Applic

-0.63

 Carth

-0.62

ukong

-0.62

 âľ

-0.61

POSITIVE LOGITS

rontal

0.71

qqa

0.67

ortion

0.66

ãĤ¦ãĤ¹

0.65

":""},{"

0.64

urat

0.63

details

0.63

rict

0.63

Hart

0.63

hov

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.