INDEX

Explanations

No Explanations Found

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

1:0.08

2:0.07

3:0.08

4:0.09

5:0.08

6:0.08

7:0.09

8:0.07

9:0.08

10:0.07

11:0.08

Negative Logits

 autistic

-2.73

 gadget

-2.73

 organising

-2.65

 organisers

-2.57

 geek

-2.54

bug

-2.41

 disabled

-2.40

 clicked

-2.38

 sticky

-2.38

 lapt

-2.37

POSITIVE LOGITS

Corp

3.00

edom

2.90

byss

2.77

Dir

2.75

breeding

2.71

DragonMagazine

2.68

Church

2.68

ngth

2.67

MpServer

2.67

arin

2.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.