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

 Sabb

-0.74

 souls

-0.69

 traveler

-0.68

tur

-0.67

 civilization

-0.66

 Beir

-0.65

Alv

-0.65

 Stir

-0.65

 hospitality

-0.64

 appell

-0.64

POSITIVE LOGITS

aucus

0.73

ilib

0.71

rust

0.69

ues

0.67

onse

0.67

DOM

0.66

 veto

0.64

culosis

0.64

 Calls

0.63

ansion

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.