INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

BI

-0.71

iple

-0.67

urgical

-0.66

omen

-0.66

zai

-0.64

iasis

-0.61

aturdays

-0.61

aceous

-0.61

 pork

-0.60

ocene

-0.60

POSITIVE LOGITS

 Wraith

0.79

 Protector

0.71

ĵĺ

0.68

 rall

0.68

mate

0.67

 Atlantis

0.67

 Cipher

0.66

\.

0.66

etheless

0.65

 Corvette

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.