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

ozy

-0.75

othe

-0.70

illation

-0.69

matic

-0.69

ickle

-0.69

Gov

-0.65

EFF

-0.65

atic

-0.64

xon

-0.64

oe

-0.64

POSITIVE LOGITS

 bonded

0.81

ĸļ

0.72

wcs

0.72

iens

0.71

 targ

0.68

vertisements

0.67

eatures

0.65

 negoti

0.64

introdu

0.63

ushi

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.