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

inventoryQuantity

-0.92

Lind

-0.80

Í

-0.76

Anonymous

-0.75

rawdownloadcloneembedreportprint

-0.71

liest

-0.69

Ler

-0.69

 Kelvin

-0.68

redients

-0.65

keys

-0.64

POSITIVE LOGITS

fax

0.71

 ally

0.69

bridge

0.66

©¶æ

0.65

raq

0.64

 captive

0.64

ridges

0.63

 exhausted

0.62

 stra

0.59

¥ŀ

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.