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

çīĪ

-0.81

 GOODMAN

-0.78

ãĥĨãĤ£

-0.74

 Longh

-0.66

ule

-0.66

ulet

-0.66

orthy

-0.65

é¾įåĸļå£«

-0.64

Accessory

-0.64

gered

-0.62

POSITIVE LOGITS

 endorsements

0.67

 whats

0.63

 pleas

0.62

aeper

0.61

wikipedia

0.61

blance

0.60

 electrons

0.60

 tweaking

0.60

 satell

0.59

luc

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.