INDEX

Explanations

mathematical notation and concepts related to set theory, in particular focusing on the Cantor diagonal argument and Cantor's theorem

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.05

1:0.04

2:0.18

3:0.14

4:0.13

5:0.10

6:0.04

7:0.03

8:0.07

9:0.07

10:0.07

11:0.03

Negative Logits

wolves

-1.26

ktop

-1.24

 progressing

-1.16

iliated

-1.16

 fracturing

-1.13

 experimenting

-1.10

 Lutheran

-1.10

 inexper

-1.09

 attrition

-1.08

POSITIVE LOGITS

�

1.46

�

1.45

Prev

1.36

�

1.36

�

1.36

rain

1.34

��

1.33

�

1.32

�

1.30

�

1.28

Activations Density 0.003%

mathematical notation and concepts related to set theory, in particular focusing on the Cantor diagonal argument and Cantor's theorem

No Known Activations