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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aturally

-0.75

renheit

-0.72

 Mehran

-0.71

opian

-0.71

orney

-0.69

iddles

-0.69

 subscrib

-0.66

itious

-0.66

lighting

-0.66

uate

-0.65

POSITIVE LOGITS

GO

0.77

VIS

0.65

 Required

0.65

ilk

0.60

 Lich

0.60

 Errors

0.59

me

0.59

kel

0.59

 Roller

0.59

MAL

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.