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

odan

-0.71

SAP

-0.65

 prosecution

-0.62

 Citation

-0.61

cod

-0.61

HL

-0.61

 referee

-0.60

 Confederation

-0.60

urat

-0.60

](

-0.59

POSITIVE LOGITS

ngth

0.91

 Vaugh

0.79

)}

0.79

iry

0.78

ickle

0.74

yrics

0.73

UFF

0.73

aires

0.68

orkshire

0.67

sbm

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.