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

hess

-0.83

 cricket

-0.72

pei

-0.67

sterdam

-0.66

ãĤ£

-0.66

 destroyer

-0.65

virt

-0.65

 manag

-0.64

 Polo

-0.64

à©

-0.62

POSITIVE LOGITS

ilight

0.65

bri

0.63

earchers

0.61

plings

0.60

 Achievement

0.60

RM

0.59

ted

0.58

astical

0.58

itage

0.57

TNT

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.