INDEX

Explanations

references to the company's operational aspects or critical statements about its activities

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

0:0.06

1:0.09

2:0.09

3:0.08

4:0.08

5:0.08

6:0.08

7:0.07

8:0.07

9:0.08

10:0.08

11:0.08

Negative Logits

Detailed

-3.06

Distance

-2.68

-2.57

�

-2.56

oren

-2.53

-2.52

 deforestation

-2.52

––

-2.48

�

-2.47

destruct

-2.46

POSITIVE LOGITS

Osw

2.92

 theme

2.56

 Jade

2.46

 advertised

2.45

 promotion

2.40

 preferably

2.32

 guest

2.30

 boon

2.28

 promo

2.27

 quotes

2.25

Activations Density 0.000%

references to the company's operational aspects or critical statements about its activities

No Known Activations