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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ulkan

-0.80

osponsors

-0.74

 intercepted

-0.72

adeon

-0.70

 radi

-0.70

Oo

-0.69

head

-0.67

 Hirosh

-0.67

psey

-0.66

utic

-0.66

POSITIVE LOGITS

Article

0.86

alde

0.74

Fund

0.72

Explore

0.71

CHA

0.71

CRE

0.69

Lie

0.68

Pros

0.68

Fram

0.67

Learn

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.