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

 induction

-0.71

TB

-0.63

chens

-0.63

ched

-0.62

 threaded

-0.62

BW

-0.61

throp

-0.61

bush

-0.60

chet

-0.60

lot

-0.60

POSITIVE LOGITS

liga

0.74

 plunder

0.72

VIDIA

0.70

ĸļ

0.68

 ruin

0.66

èª

0.65

Commerce

0.65

0.63

0000000000000000

0.62

krit

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.