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

¥µ

-0.80

ÃĽ

-0.77

ļéĨĴ

-0.71

 compat

-0.67

 consulting

-0.66

 newcom

-0.66

 anch

-0.66

ILLE

-0.65

ĻĤ

-0.65

¶æ

-0.63

POSITIVE LOGITS

 Deaths

0.86

 Emails

0.78

notations

0.78

flation

0.71

interstitial

0.69

 Valiant

0.68

luck

0.68

 Dise

0.67

aan

0.65

lah

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.