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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ÄŁ

-0.86

KO

-0.80

OGR

-0.72

zbollah

-0.71

RIP

-0.68

ublic

-0.67

terday

-0.64

È

-0.63

hra

-0.63

SOURCE

-0.63

POSITIVE LOGITS

 Ghosts

0.67

ÃŃa

0.65

amation

0.64

mouth

0.62

ansas

0.62

 foot

0.61

Initialized

0.61

iosity

0.61

 Marriott

0.60

bred

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.