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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

roma

-0.81

acco

-0.77

eah

-0.76

ourn

-0.72

 Thrones

-0.70

ioned

-0.68

ismo

-0.68

akeru

-0.66

 Saud

-0.66

rero

-0.65

POSITIVE LOGITS

inventoryQuantity

0.82

Frameworks

0.81

TPPStreamerBot

0.77

rpm

0.72

rss

0.71

Lear

0.70

Dick

0.69

soDeliveryDate

0.68

ITNESS

0.68

Resources

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.