INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ioned

-0.86

otte

-0.77

olin

-0.71

 bene

-0.67

patient

-0.63

usat

-0.63

 Huntington

-0.63

ingham

-0.63

agnetic

-0.62

orld

-0.61

POSITIVE LOGITS

 proble

0.87

 sacrific

0.71

 vulner

0.71

 SHOULD

0.70

 livest

0.69

conservancy

0.69

ãĥ¼ãĥ³

0.68

dayName

0.66

 mosqu

0.66

 Bastard

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.