INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

nikov

-0.75

umber

-0.73

atorium

-0.69

[_

-0.66

roo

-0.65

schild

-0.64

 peas

-0.63

dk

-0.63

ldom

-0.62

reditary

-0.62

POSITIVE LOGITS

Pes

0.76

pes

0.72

thumbnails

0.69

illin

0.66

 Sachs

0.65

Pal

0.63

Pie

0.62

FORM

0.62

Cec

0.62

 Polaris

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.