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

Negative Logits

 Brennan

-0.76

 Kaine

-0.73

 Bore

-0.73

rones

-0.72

geries

-0.71

rers

-0.70

 Sutherland

-0.70

 Bundy

-0.69

uda

-0.67

stop

-0.66

POSITIVE LOGITS

soDeliveryDate

0.77

essen

0.77

 mortg

0.73

 partName

0.68

bleacher

0.67

 outwe

0.64

 cryptoc

0.63

 endorsements

0.61

iac

0.60

 induct

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.