INDEX

Explanations

references to specific entertainment companies or services

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

Not in Any Lists

No Comments

Head Attr Weights

0:0.07

1:0.03

2:0.05

3:0.06

4:0.04

5:0.04

6:0.21

7:0.06

8:0.05

9:0.26

10:0.04

11:0.04

Negative Logits

 Patch

-3.31

Zer

-3.30

Tracker

-3.29

git

-3.24

 Gibbs

-3.24

 Sigma

-3.23

 Sche

-3.15

 patch

-3.15

Kal

-3.12

 bike

-3.11

POSITIVE LOGITS

tainment

7.23

 Entertainment

7.06

 entertainment

6.47

 entertain

5.02

 Enix

4.02

 ENTER

3.91

 entertained

3.78

advertising

3.69

 Gaming

3.64

 entert

3.61

Activations Density 0.000%

references to specific entertainment companies or services

No Comments

No Known Activations