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

uilt

-0.80

 Saud

-0.73

onement

-0.71

eworld

-0.67

 liner

-0.67

cients

-0.64

letcher

-0.63

estinal

-0.63

rican

-0.62

 outfield

-0.61

POSITIVE LOGITS

illin

0.79

DragonMagazine

0.76

Next

0.68

otti

0.67

ogo

0.66

Els

0.65

 Medals

0.65

zip

0.65

EVs

0.64

abs

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.