INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

allery

-0.81

ipal

-0.69

 Adin

-0.69

inventoryQuantity

-0.68

rison

-0.67

riet

-0.65

inion

-0.63

veyard

-0.62

 redundancy

-0.61

ourke

-0.61

POSITIVE LOGITS

verning

0.73

ept

0.67

 Skies

0.66

 Gaul

0.66

juries

0.65

ãĥīãĥ©ãĤ´ãĥ³

0.65

ynes

0.64

PsyNetMessage

0.63

other

0.63

awed

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.