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

 obser

-0.67

 inexper

-0.66

-----------

-0.64

ECA

-0.61

unia

-0.61

uez

-0.58

iculty

-0.58

 Lauder

-0.57

phies

-0.57

 Barnett

-0.57

POSITIVE LOGITS

atively

0.68

joy

0.65

ifles

0.64

ware

0.63

£ı

0.61

 Dwell

0.59

ilt

0.59

pipe

0.59

lessly

0.59

ately

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.