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

Not in Any Lists

No Comments

Negative Logits

 reperto

-0.79

ties

-0.76

ebus

-0.70

rose

-0.70

etheless

-0.70

bun

-0.67

angles

-0.67

swick

-0.67

ipel

-0.66

isol

-0.65

POSITIVE LOGITS

 referendum

0.71

 Teresa

0.66

 vanity

0.64

myra

0.62

Cry

0.62

mA

0.62

 Debt

0.61

0.60

 deeds

0.59

EGIN

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.