INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Woody

-0.72

het

-0.71

rette

-0.71

 Hawth

-0.67

 Handler

-0.65

scape

-0.64

ttes

-0.63

 Cele

-0.63

ultural

-0.63

 Sunset

-0.62

POSITIVE LOGITS

merce

0.71

izu

0.68

BIP

0.68

 constitu

0.68

RT

0.68

vP

0.64

mith

0.64

Pie

0.64

oled

0.64

ategory

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.