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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Independence

-0.75

 totality

-0.67

 Americ

-0.65

'"

-0.64

 Craigslist

-0.63

conservancy

-0.62

 Calais

-0.61

 nucleus

-0.60

 SetTextColor

-0.60

 Pavilion

-0.60

POSITIVE LOGITS

yss

0.78

rish

0.72

choes

0.69

vous

0.69

ould

0.69

oyal

0.66

imaru

0.66

emonium

0.64

appa

0.64

icking

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.