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

 Sins

-0.72

 Mysteries

-0.69

 Osiris

-0.69

Dul

-0.65

 Species

-0.65

 Taxes

-0.64

 Roots

-0.64

 Values

-0.63

 Responsibility

-0.63

 Position

-0.62

POSITIVE LOGITS

cher

0.87

lear

0.73

cast

0.71

warn

0.71

iot

0.70

chers

0.68

ipher

0.67

saw

0.66

iper

0.66

pel

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.