INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 grace

-0.68

 Dinosaur

-0.67

sic

-0.66

weather

-0.65

cci

-0.62

pton

-0.62

ulla

-0.60

Clar

-0.60

how

-0.60

 pregn

-0.59

POSITIVE LOGITS

TB

0.76

ilib

0.72

hemy

0.72

ovie

0.69

 paraly

0.68

oms

0.64

ilon

0.63

ovies

0.63

ctors

0.60

*/(

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.