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

 Destination

-0.84

ocation

-0.83

runs

-0.82

 IMAGES

-0.73

busters

-0.72

hops

-0.70

=]

-0.68

Interstitial

-0.68

enegger

-0.68

buster

-0.68

POSITIVE LOGITS

 Aden

0.74

Chan

0.74

 fluoride

0.72

 millenn

0.70

aiman

0.67

wcsstore

0.67

 Hansen

0.65

 Chan

0.63

 Dent

0.63

 Seym

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.