INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

çīĪ

-0.91

iHUD

-0.73

":"/

-0.71

orah

-0.69

 Forgotten

-0.68

illac

-0.66

HTTP

-0.65

 UNCLASSIFIED

-0.63

haus

-0.62

ocrine

-0.59

POSITIVE LOGITS

iencies

0.74

dayName

0.70

retty

0.70

 cryst

0.67

 caut

0.66

iott

0.65

apsed

0.65

 Jays

0.62

lied

0.62

swick

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.