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

oqu

-0.71

rogram

-0.71

mson

-0.71

 Sheldon

-0.70

RAL

-0.69

advertisement

-0.68

ioxide

-0.68

vich

-0.66

assador

-0.65

vernment

-0.64

POSITIVE LOGITS

 lapt

0.75

 confessions

0.70

 belongs

0.68

 comr

0.68

 persever

0.68

 confir

0.66

 tendencies

0.65

Relations

0.64

 chops

0.62

 exceed

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.