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

 pedigree

-0.67

 airst

-0.66

 slick

-0.65

 horizont

-0.64

 pane

-0.64

 installer

-0.63

iltration

-0.63

 Towns

-0.63

antry

-0.62

Tx

-0.61

POSITIVE LOGITS

harm

0.72

 recl

0.71

olit

0.71

 revol

0.70

bear

0.70

va

0.68

chest

0.68

esson

0.68

respond

0.68

compl

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.