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

 stake

-0.70

 neighbour

-0.68

 neighbours

-0.68

 indul

-0.63

 royalty

-0.63

lur

-0.63

azi

-0.62

 classmate

-0.61

 favourite

-0.61

Gil

-0.61

POSITIVE LOGITS

ILA

0.85

asive

0.80

yrus

0.79

uman

0.68

 WITHOUT

0.68

externalActionCode

0.68

enthal

0.67

ulus

0.66

haus

0.66

 Synd

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.