INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 RAND

-0.74

OY

-0.74

FRI

-0.72

ergic

-0.70

GC

-0.69

server

-0.67

Notice

-0.66

 Cosmic

-0.66

few

-0.66

Î³

-0.65

POSITIVE LOGITS

ascript

0.72

ongevity

0.71

icut

0.68

lev

0.64

 abol

0.64

 vomit

0.63

uala

0.63

 lapt

0.62

 advoc

0.62

unia

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.