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

tainment

-0.84

perty

-0.76

dropping

-0.70

zai

-0.69

ISIS

-0.68

staking

-0.67

cks

-0.66

dro

-0.65

hip

-0.65

DragonMagazine

-0.64

POSITIVE LOGITS

0.71

olas

0.70

gio

0.66

 Thames

0.64

CF

0.63

adden

0.63

 Durham

0.62

lex

0.61

opia

0.61

urion

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.