INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

arations

-0.73

xes

-0.72

 phased

-0.67

Xia

-0.66

 Skydragon

-0.66

VIS

-0.62

odium

-0.61

 Disabled

-0.61

orph

-0.61

 Paraly

-0.60

POSITIVE LOGITS

LLOW

0.67

ANN

0.67

Owner

0.66

catentry

0.65

NB

0.64

inventoryQuantity

0.64

Ü

0.63

Ö¼

0.63

anut

0.62

abc

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.