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

served

-0.78

ople

-0.76

Lew

-0.73

 furnished

-0.71

 Tradable

-0.69

shown

-0.68

fixed

-0.67

urbed

-0.65

 Printed

-0.64

placed

-0.63

POSITIVE LOGITS

cules

0.79

OHN

0.69

ĸļ

0.69

 ethn

0.68

 flashlight

0.68

racuse

0.68

ategy

0.66

idas

0.64

 nostalg

0.62

htaking

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.