INDEX

Explanations

cooking instructions and procedures in recipes

New Auto-Interp

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

1:0.02

2:0.05

3:0.10

4:0.04

5:0.06

6:0.02

7:0.04

8:0.04

9:0.13

10:0.26

11:0.15

Negative Logits

 dating

-1.20

ivism

-1.11

redit

-1.10

 prostitute

-1.10

 motivational

-1.10

 prostitutes

-1.08

 academ

-1.07

 autobi

-1.06

 campaigning

-1.06

 slang

-1.05

POSITIVE LOGITS

++)

1.37

 Instructions

1.36

Shape

1.34

>]

1.29

Optional

1.23

』

1.19

TPPStreamerBot

1.19

}:

1.17

bsp

1.14

rawdownloadcloneembedreportprint

1.14

Activations Density 0.180%

cooking instructions and procedures in recipes

No Known Activations