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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Walters

-0.73

ARP

-0.68

NER

-0.63

auer

-0.62

ulton

-0.61

ette

-0.61

LOS

-0.61

WS

-0.60

PRES

-0.59

 abusers

-0.59

POSITIVE LOGITS

 ç¥ŀ

0.81

isphere

0.81

cgi

0.77

icated

0.74

selves

0.72

 Joined

0.70

romeda

0.68

)</

0.67

 Bulgar

0.66

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.