INDEX

Explanations

references to directors, particularly their achievements and contributions in animation, especially within Pixar films

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

0:0.27

1:0.05

2:0.07

3:0.04

4:0.07

5:0.02

6:0.06

7:0.12

8:0.02

9:0.06

10:0.12

11:0.05

Negative Logits

 Georgian

-4.23

 Ashe

-3.92

 Celt

-3.89

 doping

-3.66

Aeg

-3.37

assium

-3.36

Kal

-3.31

ovych

-3.26

enko

-3.26

 tavern

-3.21

POSITIVE LOGITS

 Pixar

7.86

Toy

4.84

 Doct

4.61

Toy

4.50

 Cars

3.73

 Dinosaur

3.69

 Woody

3.66

Coy

3.56

 anim

3.38

 Pupp

3.33

Activations Density 0.003%

references to directors, particularly their achievements and contributions in animation, especially within Pixar films

No Known Activations