INDEX

Explanations

specific references to troop numbers and military presence

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

0:0.09

1:0.05

2:0.07

3:0.08

4:0.09

5:0.04

6:0.12

7:0.13

8:0.03

9:0.08

10:0.04

11:0.12

Negative Logits

 Yelp

-2.79

Glass

-2.73

 Wallet

-2.64

 Blockchain

-2.58

 Owners

-2.54

Apps

-2.52

ICO

-2.52

◼

-2.48

Android

-2.46

 Youtube

-2.43

POSITIVE LOGITS

 troop

4.25

 battalion

3.84

 troops

3.65

 conting

3.60

 Afghan

3.43

 Mali

3.40

 Ethiop

3.37

 contingent

3.29

 Marines

3.29

 contingency

3.21

Activations Density 0.034%

specific references to troop numbers and military presence

No Known Activations