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

2:0.08

3:0.06

4:0.08

5:0.07

6:0.08

7:0.09

8:0.07

9:0.08

10:0.09

11:0.09

Negative Logits

 dehyd

-3.27

jing

-2.58

 UNESCO

-2.55

 compost

-2.50

 Waste

-2.44

Cola

-2.44

 doct

-2.41

 rubbish

-2.40

 ISIL

-2.33

 tint

-2.32

POSITIVE LOGITS

iability

2.65

rosso

2.59

agos

2.53

volt

2.52

 knots

2.38

azines

2.37

��

2.30

perm

2.25

ouses

2.23

 COUN

2.21

Activations Density 0.000%

No Known Activations

This feature has no known activations.