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

taboola

-0.71

WARD

-0.68

 Stephenson

-0.65

 Standing

-0.64

Attempt

-0.63

 Sitting

-0.63

Board

-0.62

adin

-0.61

 Vide

-0.61

POSITIVE LOGITS

uces

0.88

esty

0.80

uce

0.74

 Britann

0.73

ellation

0.72

ython

0.71

plates

0.71

ulence

0.70

ugg

0.70

amate

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.