INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ional

-0.77

otor

-0.76

naissance

-0.73

cycl

-0.73

ciples

-0.72

ework

-0.69

essor

-0.69

mble

-0.68

lan

-0.67

 disapp

-0.67

POSITIVE LOGITS

ahime

0.76

CLASSIFIED

0.68

CVE

0.66

IJ

0.66

 msec

0.65

ICH

0.64

 Caucasus

0.63

FLAG

0.63

OUGH

0.62

FU

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.