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

 denial

-0.67

 accrued

-0.67

getting

-0.66

 malf

-0.65

had

-0.63

 lingering

-0.62

Poo

-0.62

 disparity

-0.61

aston

-0.61

 manif

-0.61

POSITIVE LOGITS

Square

0.74

ãĤ§

0.73

Bi

0.71

 Scandinavian

0.71

oto

0.71

Þ

0.71

zsche

0.70

äºĶ

0.69

LV

0.68

oho

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.