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

Pg

-0.70

hover

-0.70

 suffice

-0.68

 opacity

-0.64

)}

-0.64

ffer

-0.63

Equ

-0.62

areth

-0.62

 nonetheless

-0.61

eq

-0.61

POSITIVE LOGITS

urse

0.70

yssey

0.69

ivals

0.65

 Circuit

0.65

 Aboriginal

0.65

opal

0.64

nings

0.64

RTX

0.64

 contracting

0.63

paralle

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.