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

arag

-0.76

algia

-0.69

apolis

-0.69

redo

-0.67

 Reconstruction

-0.67

luaj

-0.66

ijah

-0.66

 Guatem

-0.62

 Commerce

-0.61

 Justice

-0.61

POSITIVE LOGITS

idences

1.02

IDENT

0.71

flix

0.69

olver

0.68

faced

0.66

daq

0.65

dit

0.65

parent

0.64

mut

0.64

idon

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.