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

 Companies

-0.68

CLA

-0.67

 Sons

-0.63

}}}

-0.62

UFC

-0.60

 Malcolm

-0.60

amon

-0.59

 Lisp

-0.59

Ð´

-0.59

 Recall

-0.59

POSITIVE LOGITS

eteenth

0.78

ger

0.75

ľ

0.73

anqu

0.73

rer

0.71

vernment

0.71

ciating

0.70

 Nether

0.70

veyard

0.69

uper

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.