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

utical

-0.77

ially

-0.70

 Palestin

-0.67

grain

-0.65

 Aren

-0.63

udo

-0.62

 sparing

-0.62

uits

-0.61

Protect

-0.61

tarians

-0.60

POSITIVE LOGITS

],

0.69

ticket

0.66

 crow

0.65

)]

0.65

âĸº

0.62

]-

0.61

 traveller

0.61

 aster

0.60

%]

0.60

digit

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.