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

ingly

-0.73

backs

-0.70

ISO

-0.68

Gate

-0.68

aries

-0.62

ohan

-0.61

achi

-0.61

back

-0.61

aki

-0.61

udo

-0.60

POSITIVE LOGITS

cffff

0.98

ufact

0.89

 WATCHED

0.82

forming

0.76

Ń·

0.75

é¾įåĸļå£«

0.74

¥µ

0.74

 VIDEOS

0.73

anmar

0.73

aughs

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.