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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bats

-0.85

FactoryReloaded

-0.71

worm

-0.68

STDOUT

-0.68

worms

-0.67

 expansions

-0.66

Marginal

-0.65

 Myster

-0.64

backer

-0.64

PDATE

-0.63

POSITIVE LOGITS

urdy

0.75

 grasp

0.70

itual

0.69

ishy

0.68

heit

0.67

ritional

0.64

orthy

0.64

 Vlad

0.64

 Swim

0.63

det

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.