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

FK

-0.69

pins

-0.66

pin

-0.66

interstitial

-0.66

Pil

-0.66

 Tonight

-0.63

pac

-0.62

gear

-0.61

pull

-0.60

mos

-0.58

POSITIVE LOGITS

eme

0.80

emale

0.77

yrinth

0.69

hent

0.68

ahime

0.68

emonium

0.67

iq

0.66

hement

0.65

eph

0.64

 sqor

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.