INDEX

Explanations

commands and package management terms related to software installation on Linux systems

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

0:0.04

1:0.04

2:0.11

3:0.04

4:0.10

5:0.14

6:0.04

7:0.10

8:0.06

9:0.11

10:0.06

11:0.09

Negative Logits

.""

-3.15

!?"

-3.10

."[

-2.99

?".

-2.86

!".

-2.82

!!"

-2.77

?",

-2.76

!",

-2.69

?!"

-2.66

.",

-2.65

POSITIVE LOGITS

 python

4.07

 sudo

4.06

cd

3.03

sudo

3.00

pip

2.99

python

2.95

lib

2.94

git

2.92

gol

2.82

py

2.68

Activations Density 0.005%

commands and package management terms related to software installation on Linux systems

No Known Activations