INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

trak

-0.76

iral

-0.73

iversal

-0.68

hai

-0.68

rical

-0.67

umin

-0.67

 exited

-0.65

 disemb

-0.63

eca

-0.63

animous

-0.63

POSITIVE LOGITS

ãĤ§

0.73

arette

0.64

orce

0.63

ãĥ³ãĤ¸

0.63

 Mechdragon

0.62

 peach

0.60

 Adobe

0.58

crop

0.57

anton

0.57

 Grimoire

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.