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

0:0.09

1:0.07

2:0.08

3:0.08

4:0.08

5:0.09

6:0.08

7:0.07

8:0.07

9:0.07

10:0.07

11:0.08

Negative Logits

 merchandise

-2.62

abilia

-2.61

 merch

-2.45

 magnets

-2.34

ograph

-2.33

sho

-2.31

 sellers

-2.30

stasy

-2.29

 taxable

-2.29

\/

-2.29

POSITIVE LOGITS

oglu

2.94

Erd

2.75

 Operations

2.59

Campaign

2.57

 Diplom

2.51

 Strateg

2.49

 Commands

2.49

 Fired

2.48

oğ

2.48

 Campaign

2.48

Activations Density 0.000%

No Known Activations

This feature has no known activations.