INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Franc

-0.80

 Crown

-0.67

 Carbuncle

-0.67

 Berry

-0.66

 gorge

-0.65

 Chop

-0.65

 Gorge

-0.64

 Seed

-0.63

 Sakuya

-0.63

ordinate

-0.62

POSITIVE LOGITS

UFC

0.77

omsky

0.75

osite

0.73

alions

0.72

alan

0.71

opol

0.70

kinson

0.69

jriwal

0.68

oval

0.68

axis

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.