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

 Â»

-0.68

married

-0.64

 Malk

-0.62

dro

-0.62

 deposition

-0.61

irgin

-0.61

oun

-0.60

earchers

-0.60

 Stain

-0.59

onda

-0.58

POSITIVE LOGITS

Dial

0.93

ä½¿

0.81

Crew

0.80

entimes

0.77

Adapter

0.75

TAG

0.73

Wars

0.73

Oracle

0.73

Else

0.72

Writer

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.