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

Initialized

-0.75

missions

-0.75

Bear

-0.74

 Earthqu

-0.72

McA

-0.70

IJ

-0.70

Operation

-0.69

âĹ¼

-0.69

Dur

-0.68

 Kirin

-0.68

POSITIVE LOGITS

xual

0.89

orial

0.68

 silent

0.67

 hinge

0.67

iple

0.66

 ribbon

0.66

 succinct

0.64

 ubiquitous

0.64

 broad

0.63

 discriminated

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.