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

itudes

-0.83

itud

-0.75

soDeliveryDate

-0.71

auc

-0.70

erous

-0.66

 worsh

-0.66

 ministries

-0.65

jah

-0.64

ocated

-0.63

anism

-0.63

POSITIVE LOGITS

VIEW

0.72

elli

0.63

Cu

0.63

PLA

0.63

 Garry

0.60

Ford

0.60

mare

0.59

culus

0.59

utor

0.59

Maw

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.