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

lishes

-0.74

CFL

-0.74

upd

-0.70

 embed

-0.70

 Zombies

-0.68

 Burk

-0.66

embed

-0.63

neau

-0.63

 compress

-0.62

iaz

-0.62

POSITIVE LOGITS

ategory

0.68

ĪĴ

0.67

irs

0.67

arent

0.66

erva

0.64

 Heritage

0.62

eryl

0.61

angered

0.61

sequ

0.61

bered

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.