INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 trough

-0.68

IOR

-0.65

 imag

-0.65

Factor

-0.64

 deform

-0.60

DM

-0.59

 reception

-0.59

 Outbreak

-0.59

dm

-0.58

 Goodman

-0.58

POSITIVE LOGITS

1.13

ĺħ

0.78

inventoryQuantity

0.78

NAV

0.76

bryce

0.75

ĪĴ

0.73

 Cosponsors

0.70

leck

0.69

catentry

0.69

ivot

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.