INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ħĭ

-0.87

Ħ¢

-0.75

gradient

-0.66

 moons

-0.66

undo

-0.65

obal

-0.64

 Windsor

-0.62

tle

-0.61

 Guan

-0.61

Eco

-0.61

POSITIVE LOGITS

ecause

0.69

usercontent

0.67

ahoo

0.64

 WATCHED

0.63

 prescribe

0.63

afety

0.63

Spoiler

0.62

EEE

0.62

iculty

0.62

glers

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.