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

)]

-0.87

å°Ĩ

-0.77

LCS

-0.76

ELS

-0.76

 guiIcon

-0.71

çļĦ

-0.68

ADS

-0.68

});

-0.68

NRS

-0.67

Dou

-0.66

POSITIVE LOGITS

 Kafka

0.71

bestos

0.71

wagen

0.71

 Wander

0.70

 Swordsman

0.69

hander

0.68

 passage

0.66

mage

0.65

eness

0.64

gaard

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.