INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 opio

-0.72

intendo

-0.69

 contrace

-0.67

 lender

-0.67

ivo

-0.67

 estab

-0.66

 redes

-0.66

 Bots

-0.66

 joystick

-0.65

 Bras

-0.64

POSITIVE LOGITS

umn

0.73

)]

0.66

Raw

0.65

scientific

0.65

ames

0.64

Claim

0.64

chemy

0.64

Sche

0.64

Que

0.63

Labor

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.