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

 Bane

-0.87

 Trance

-0.79

 Doomsday

-0.78

 Hannity

-0.71

 Hades

-0.70

 Limbaugh

-0.70

 Ares

-0.69

Wyr

-0.68

 Elves

-0.67

 Warlock

-0.64

POSITIVE LOGITS

DN

0.86

:[

0.81

eny

0.76

çīĪ

0.75

imen

0.73

ynchron

0.70

pire

0.69

ohl

0.69

cher

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.