INDEX

Explanations

terms related to programming concepts involving reference parameters

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.08

1:0.07

2:0.10

3:0.09

4:0.08

5:0.08

6:0.07

7:0.07

8:0.07

9:0.07

10:0.08

11:0.08

Negative Logits

YC

-3.43

 Consent

-2.88

stocks

-2.63

••

-2.62

cca

-2.52

phan

-2.52

/,

-2.48

®,

-2.47

JUST

-2.42

breeding

-2.41

POSITIVE LOGITS

�

2.58

dinand

2.47

 ende

2.41

 whistleblower

2.29

enge

2.27

 veteran

2.25

 laure

2.20

 slain

2.20

htaking

2.19

ubis

2.17

Activations Density 0.000%

terms related to programming concepts involving reference parameters

No Known Activations