INDEX

Explanations

names related to notable figures or entities

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.10.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.16

1:0.20

2:0.02

3:0.04

4:0.02

5:0.01

6:0.10

7:0.11

8:0.04

9:0.03

10:0.15

11:0.07

Negative Logits

Yin

-3.05

Miy

-2.89

 yoga

-2.79

 Nanto

-2.65

 Yuan

-2.64

Hua

-2.63

 Shin

-2.63

Kyl

-2.60

Sak

-2.60

 Suzuki

-2.54

POSITIVE LOGITS

ennis

3.06

arge

3.05

EngineDebug

2.99

icket

2.95

Redd

2.81

minecraft

2.77

ett

2.70

arrow

2.64

urn

2.62

udeb

2.60

Activations Density 0.065%

names related to notable figures or entities

No Known Activations