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

RTX

-0.82

 Bucket

-0.75

EMBER

-0.73

ABE

-0.71

ITH

-0.69

 Prediction

-0.66

Buyable

-0.66

 Dying

-0.64

âģ

-0.64

ACP

-0.63

POSITIVE LOGITS

achev

0.75

esty

0.74

addr

0.71

unia

0.69

undai

0.69

utral

0.68

 paran

0.66

 sweets

0.65

humane

0.64

illance

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.