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

 critically

-0.78

 constitu

-0.78

aughed

-0.72

acus

-0.69

aminer

-0.68

romeda

-0.68

aukee

-0.67

incinn

-0.66

riter

-0.65

ratulations

-0.65

POSITIVE LOGITS

gap

0.71

Ì¶

0.69

fet

0.65

window

0.65

 displays

0.63

 headsets

0.62

Zone

0.62

endered

0.61

ahime

0.61

 pores

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.