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

krit

-0.90

uran

-0.74

ulator

-0.69

akra

-0.62

ulators

-0.61

acea

-0.61

kj

-0.61

Ì¶

-0.60

:{

-0.60

 proble

-0.60

POSITIVE LOGITS

UX

0.78

CTV

0.74

Pont

0.74

Cooldown

0.74

Sensor

0.73

Files

0.73

Anth

0.71

taboola

0.67

vous

0.67

EAR

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.