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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

onym

-0.69

 Guards

-0.68

racted

-0.66

ngth

-0.66

ife

-0.65

rov

-0.65

achus

-0.65

HL

-0.65

ope

-0.65

 compuls

-0.64

POSITIVE LOGITS

 Fiesta

0.69

updated

0.67

 variation

0.67

TOR

0.65

 Referred

0.65

ardless

0.65

invoke

0.65

present

0.64

[];

0.64

ado

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.