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

UTF

-0.87

 Dunham

-0.77

 Finch

-0.73

staking

-0.68

66666666

-0.65

ansson

-0.65

 Angus

-0.64

zilla

-0.64

acters

-0.64

Tud

-0.63

POSITIVE LOGITS

 Masquerade

0.77

 carrier

0.69

 intention

0.69

 assailant

0.66

 combination

0.65

uyomi

0.64

âĻ¥

0.64

 Chocobo

0.64

 loser

0.63

 Applicant

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.