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

Not in Any Lists

No Comments

Negative Logits

isky

-0.69

rer

-0.67

athlon

-0.66

arch

-0.66

 Thick

-0.64

olic

-0.64

roud

-0.62

 Asset

-0.62

atana

-0.62

achus

-0.61

POSITIVE LOGITS

oun

0.81

 mysteries

0.77

Reviewer

0.75

comings

0.65

Zub

0.64

stories

0.64

 conflic

0.62

dies

0.60

 realism

0.60

 Moroc

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.