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

Depth

-0.71

winter

-0.64

¿½

-0.63

jew

-0.63

umerable

-0.63

HUD

-0.60

irth

-0.60

Ka

-0.59

DAQ

-0.59

 Beaut

-0.58

POSITIVE LOGITS

 anarchism

0.69

oulos

0.68

istani

0.67

gow

0.61

 Ecuador

0.59

 comrade

0.59

 bent

0.59

abo

0.58

 Rwanda

0.58

 ante

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.