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

ULE

-0.86

 Malk

-0.86

entry

-0.84

MK

-0.75

Shut

-0.69

enter

-0.66

 Roose

-0.66

owe

-0.64

olt

-0.63

JFK

-0.63

POSITIVE LOGITS

 depos

0.82

ciating

0.78

imon

0.70

ths

0.70

cript

0.69

 handwriting

0.68

cycl

0.67

ihara

0.66

DNA

0.66

 affidav

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.