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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

htar

-0.76

oscope

-0.75

Reviewer

-0.75

Loading

-0.70

jri

-0.70

Ire

-0.68

BALL

-0.68

TAG

-0.67

Neh

-0.66

Pand

-0.66

POSITIVE LOGITS

 citiz

0.89

©¶æ

0.75

activity

0.73

atures

0.70

theless

0.70

akers

0.69

 facult

0.66

 excellence

0.66

 juven

0.65

compliance

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.