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

closure

-0.82

aper

-0.81

rose

-0.76

appa

-0.75

rider

-0.73

-0.72

coe

-0.72

pass

-0.71

ector

-0.70

erest

-0.70

POSITIVE LOGITS

SJ

0.66

 Monstrous

0.65

 Hiroshima

0.65

 redes

0.64

 slender

0.64

 Sylvia

0.61

 Witches

0.60

 squared

0.60

 sake

0.60

 spiritually

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.