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

 tacit

-0.74

 prevailing

-0.74

 volunt

-0.71

nud

-0.70

idon

-0.66

wcs

-0.66

 bets

-0.66

 IMAGES

-0.66

 prejudices

-0.66

roots

-0.65

POSITIVE LOGITS

Bee

0.71

 Frontier

0.69

Tech

0.68

Stre

0.66

 Airl

0.66

Anth

0.65

 Canada

0.65

 Turk

0.64

Oak

0.63

Tur

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.