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

0:0.08

1:0.09

2:0.07

3:0.07

4:0.07

5:0.08

6:0.08

7:0.09

8:0.08

9:0.08

10:0.07

11:0.08

Negative Logits

ucket

-3.43

 Toby

-2.90

 Dele

-2.85

=-=-=-=-=-=-=-=-

-2.81

 Rodrigo

-2.79

eki

-2.77

hoe

-2.71

boss

-2.70

 Pablo

-2.67

 Duncan

-2.66

POSITIVE LOGITS

atin

2.56

 Genetics

2.55

Syn

2.48

 Generations

2.45

mit

2.44

 radi

2.43

 Dian

2.40

 antigen

2.40

fam

2.39

 blush

2.36

Activations Density 0.000%

No Known Activations

This feature has no known activations.