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

negie

-0.76

 mathemat

-0.71

croft

-0.71

irmation

-0.68

Glass

-0.66

 guiActiveUnfocused

-0.66

wald

-0.65

 resumed

-0.64

soDeliveryDate

-0.63

ronic

-0.63

POSITIVE LOGITS

otypes

0.77

otype

0.75

icals

0.73

ically

0.72

gments

0.69

issues

0.68

imp

0.65

))))

0.63

oots

0.62

icable

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.