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

 scrim

-0.71

 elim

-0.70

 myster

-0.70

GW

-0.70

ertation

-0.70

krit

-0.70

ĪĴ

-0.69

Offline

-0.66

iggurat

-0.66

uti

-0.65

POSITIVE LOGITS

 Nero

0.83

oses

0.79

iannopoulos

0.75

Ce

0.68

 Scal

0.66

illary

0.66

gow

0.65

 Xavier

0.65

flow

0.65

umbers

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.