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

cies

-0.92

ocide

-0.78

ions

-0.75

heit

-0.74

grave

-0.74

isites

-0.73

iage

-0.73

hess

-0.72

igham

-0.69

 religions

-0.68

POSITIVE LOGITS

Peb

0.70

rig

0.64

 Duty

0.62

âĸº

0.62

 scrimmage

0.61

Pok

0.61

 Medal

0.61

 Corpus

0.59

 Skull

0.59

DERR

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.