INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

leneck

-0.78

 Ambro

-0.74

cific

-0.71

abies

-0.68

udi

-0.68

inka

-0.67

nox

-0.66

daq

-0.66

idency

-0.63

ud

-0.63

POSITIVE LOGITS

hower

0.76

Magikarp

0.70

RAW

0.67

0.66

 predic

0.65

 caches

0.64

GMT

0.64

 seiz

0.62

è¦ļéĨĴ

0.60

ãĥĩ

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.