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

raints

-0.82

ibur

-0.80

sis

-0.79

nces

-0.78

aque

-0.78

zeb

-0.75

antage

-0.74

ibly

-0.74

anke

-0.74

BuyableInstoreAndOnline

-0.74

POSITIVE LOGITS

 association

0.72

 Metropolitan

0.69

 Vertical

0.65

 Organ

0.61

 distributed

0.61

 everyday

0.60

 Copyright

0.60

 Educational

0.59

Flor

0.59

Sr

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.