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

bush

-0.79

HEAD

-0.75

 Ples

-0.72

 Guinness

-0.68

heads

-0.66

head

-0.66

utm

-0.66

AMI

-0.65

 DRAGON

-0.65

IRD

-0.64

POSITIVE LOGITS

etheless

0.77

affer

0.71

cept

0.66

istan

0.64

 knack

0.64

ifference

0.64

rompt

0.64

ortunate

0.63

essim

0.63

 grace

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.