INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ãĥİ

-0.85

è£ħ

-0.82

ãĥ¤

-0.75

AIN

-0.74

ITNESS

-0.71

_>

-0.70

âķĲâķĲ

-0.70

ãĥĨãĤ£

-0.69

 Hedge

-0.67

 GOODMAN

-0.67

POSITIVE LOGITS

 Invisible

0.70

mates

0.68

etsk

0.67

mate

0.65

imilation

0.64

MSN

0.63

cdn

0.63

slave

0.63

egal

0.62

 machines

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.