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

 Print

-0.72

 Artifact

-0.70

ieri

-0.66

 Grammy

-0.64

amaru

-0.63

 Emmy

-0.60

 split

-0.59

 severed

-0.59

 numbered

-0.58

ibli

-0.58

POSITIVE LOGITS

Thor

0.78

bage

0.77

Matt

0.69

Russ

0.69

ledge

0.69

Russell

0.68

Attributes

0.68

 Sabha

0.67

mort

0.66

Statistics

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.