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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ÃĽ

-0.87

inventoryQuantity

-0.81

EngineDebug

-0.81

Magikarp

-0.78

INO

-0.76

ILLE

-0.76

è¯

-0.76

PsyNetMessage

-0.75

fw

-0.73

çľ

-0.72

POSITIVE LOGITS

 hardest

0.82

 same

0.76

rouse

0.75

 majority

0.72

 bulk

0.70

oret

0.69

 examiner

0.69

 entire

0.69

 entirety

0.69

 simplest

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.