INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

instance

-0.72

*****

-0.67

properties

-0.66

 backwards

-0.65

aterial

-0.63

 ether

-0.63

rocket

-0.62

manship

-0.62

~~~~

-0.61

able

-0.61

POSITIVE LOGITS

 Gaul

0.68

 Rica

0.66

eah

0.64

 Bris

0.63

Gam

0.61

ILA

0.61

zhou

0.61

 Zhou

0.60

 Maze

0.60

 gorge

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.