INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

[/

-0.74

orem

-0.69

ogg

-0.69

ãĤ©

-0.68

adium

-0.68

 Guides

-0.67

":[{"

-0.67

}}}

-0.66

":{"

-0.66

 Quality

-0.64

POSITIVE LOGITS

HER

0.75

NSA

0.74

immune

0.72

GROUP

0.71

LESS

0.71

upon

0.71

 immune

0.68

 sill

0.67

milo

0.66

court

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.