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

unia

-0.79

amina

-0.78

ãĤĴ

-0.74

CHAPTER

-0.73

Ļ½

-0.71

ILLE

-0.70

ARB

-0.68

 Persia

-0.67

Mb

-0.67

Ñı

-0.67

POSITIVE LOGITS

iple

0.69

 Origin

0.64

 spirited

0.64

NS

0.63

 solicit

0.63

 artif

0.62

olicited

0.61

 Advent

0.61

 Swed

0.61

earchers

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.