INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Baxter

-0.78

=-=-

-0.77

 Dull

-0.76

inas

-0.73

vc

-0.69

own

-0.66

 bark

-0.65

 john

-0.65

Īè

-0.65

cv

-0.62

POSITIVE LOGITS

 literacy

0.73

awaru

0.72

apult

0.69

nsic

0.67

 Spectre

0.66

xen

0.65

anship

0.65

ldom

0.63

address

0.62

annis

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.