INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 scapego

-0.87

 builders

-0.71

 dwelling

-0.66

 easing

-0.62

 scra

-0.62

 burner

-0.62

 abuser

-0.60

 makers

-0.60

 mastermind

-0.59

 Creator

-0.59

POSITIVE LOGITS

moderate

0.91

irrel

0.79

scenes

0.78

grim

0.77

wcsstore

0.77

aura

0.76

neck

0.76

larg

0.74

abama

0.74

lus

0.74

Activations Density 0.000%

No Known Activations

This feature has no known activations.