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

aye

-0.69

NRS

-0.69

otten

-0.65

Ń·

-0.65

meat

-0.65

ream

-0.64

ouver

-0.64

ardon

-0.64

ordon

-0.64

APD

-0.63

POSITIVE LOGITS

 artif

0.72

bestos

0.68

imar

0.67

Unix

0.67

lass

0.65

asio

0.65

 Hiroshima

0.64

 CentOS

0.63

pmwiki

0.62

regate

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.