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

 Tradable

-0.80

 Gunn

-0.74

icken

-0.73

thora

-0.66

swick

-0.65

»Ĵ

-0.62

ermott

-0.61

 psychiatrists

-0.61

herty

-0.61

@#&

-0.61

POSITIVE LOGITS

0.78

Â·Â·

0.76

worthiness

0.76

isu

0.73

vel

0.72

ebook

0.68

gorithm

0.67

photos

0.65

vel

0.65

gorith

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.