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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pag

-0.70

apo

-0.66

 dissemination

-0.64

 orphan

-0.63

 Sources

-0.63

uality

-0.63

 ashes

-0.62

onym

-0.61

 Exile

-0.60

 exhaustion

-0.60

POSITIVE LOGITS

imeters

0.79

erial

0.77

nder

0.76

uca

0.75

asuring

0.73

ischer

0.69

eatures

0.68

tery

0.68

quickShipAvailable

0.67

isl

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.