INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

1:0.07

2:0.08

3:0.08

4:0.09

5:0.08

6:0.07

7:0.08

8:0.09

9:0.07

10:0.08

11:0.08

Negative Logits

 Paran

-1.89

ean

-1.71

 Gothic

-1.67

 Material

-1.64

creen

-1.64

 Solitaire

-1.64

endor

-1.64

itatively

-1.62

/+

-1.62

uary

-1.61

POSITIVE LOGITS

NetMessage

1.98

renched

1.81

tackle

1.76

@#&

1.63

ダ

1.62

rowd

1.59

Andre

1.57

reat

1.56

 interchange

1.56

lehem

1.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.