INDEX

Explanations

numerical statistics related to sports performance metrics

New Auto-Interp

Top Features by Cosine Similarity

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

1:0.01

2:0.20

3:0.06

4:0.12

5:0.02

6:0.11

7:0.06

8:0.14

9:0.04

10:0.05

11:0.10

Negative Logits

 guiActive

-1.63

\":

-1.44

 writer

-1.23

="#

-1.23

 follow

-1.23

 reader

-1.23

 dissent

-1.22

 sweep

-1.20

 pending

-1.18

 convoy

-1.18

POSITIVE LOGITS

bos

1.60

rex

1.44

 Bian

1.41

phia

1.41

 Canary

1.36

 サーティワン

1.32

Transfer

1.28

 Safari

1.24

 Pokémon

1.23

services

1.22

Activations Density 0.001%

numerical statistics related to sports performance metrics

No Known Activations