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

efer

-0.82

ench

-0.80

gage

-0.79

UME

-0.75

laden

-0.71

roth

-0.70

look

-0.69

mage

-0.69

IX

-0.68

ensen

-0.68

POSITIVE LOGITS

 Indies

0.76

OPLE

0.74

 cath

0.67

Reviewer

0.64

iami

0.63

 Artists

0.62

FIR

0.62

 everywhere

0.61

 Polaris

0.61

 electro

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.