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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Luxembourg

-0.73

 Luxem

-0.71

bourg

-0.71

 diplom

-0.66

 milit

-0.65

 accuse

-0.64

 Shame

-0.64

lux

-0.64

Vaj

-0.62

 Shant

-0.62

POSITIVE LOGITS

sol

0.79

 generators

0.75

orsche

0.71

Loader

0.69

agar

0.69

riot

0.68

aci

0.65

---------

0.65

acion

0.64

tery

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.