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

 metic

-0.89

arij

-0.75

itton

-0.74

ETH

-0.72

Ã³n

-0.71

urchase

-0.70

TOR

-0.69

abre

-0.67

ucha

-0.66

ase

-0.66

POSITIVE LOGITS

 echoing

0.64

Haw

0.63

 symm

0.63

MI

0.62

 Kerr

0.61

Dialog

0.61

flo

0.61

 enclosed

0.60

 ranked

0.60

hematic

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.