INDEX

Explanations

themes related to dryness and dehydration in various contexts

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_ds24576_bs4096_dc1.00e-06_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.3.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.03

1:0.02

2:0.07

3:0.19

4:0.02

5:0.03

6:0.12

7:0.16

8:0.04

9:0.08

10:0.07

11:0.12

Negative Logits

 Cosponsors

-1.51

�士

-1.41

�

-1.38

�

-1.37

UA

-1.34

atown

-1.33

vernment

-1.32

lez

-1.32

イ

-1.24

Spons

-1.23

POSITIVE LOGITS

 diesel

1.30

 compressed

1.19

 corpus

1.18

 Corpus

1.17

 decom

1.10

 decomp

1.10

 drying

1.09

 compress

1.09

 spell

1.09

 gland

1.08

Activations Density 0.010%

themes related to dryness and dehydration in various contexts

No Known Activations