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

fur

-0.74

ollar

-0.72

osher

-0.72

£ı

-0.71

nutrition

-0.71

 Tail

-0.70

ĸļ

-0.70

oured

-0.69

repair

-0.69

odynamic

-0.69

POSITIVE LOGITS

 Borders

0.72

 aloud

0.69

CES

0.67

 Instruments

0.67

 torches

0.66

 apost

0.65

 overheard

0.63

 Compass

0.62

 coasts

0.62

 swall

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.