INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 ingred

-0.70

 vind

-0.68

 aloud

-0.63

nen

-0.63

 outl

-0.63

ibilities

-0.62

 Siber

-0.61

 Publishers

-0.61

é¾įå¥ĳå£«

-0.61

aspers

-0.60

POSITIVE LOGITS

hover

0.78

aughs

0.74

 Proposition

0.73

ongo

0.73

MRI

0.69

 Sabha

0.69

LOCK

0.69

fac

0.69

umn

0.68

UI

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.