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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 unfolded

-0.73

orf

-0.72

 anat

-0.70

upper

-0.63

zee

-0.61

yna

-0.61

apa

-0.61

tymology

-0.60

GROUND

-0.60

 Allaah

-0.59

POSITIVE LOGITS

\":

0.68

hya

0.65

 Heist

0.63

¶

0.63

 Tribe

0.62

 overfl

0.62

neum

0.62

ahar

0.61

 Horde

0.60

hd

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.