INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

iety

-0.85

present

-0.83

pees

-0.80

zee

-0.79

berman

-0.78

unity

-0.74

 Wyatt

-0.72

anya

-0.71

ancock

-0.69

asca

-0.68

POSITIVE LOGITS

 reflections

0.65

¿½

0.62

Īè

0.62

oslav

0.62

 UNCLASSIFIED

0.61

 Brexit

0.59

 Toro

0.59

 Noct

0.58

 Fukushima

0.57

 Asus

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.