INDEX

Explanations

No Explanations Found

New Auto-Interp

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

0:0.07

1:0.07

2:0.10

3:0.10

4:0.07

5:0.08

6:0.06

7:0.05

8:0.08

9:0.09

10:0.09

11:0.08

Negative Logits

ystem

-1.74

eu

-1.70

lower

-1.63

ension

-1.60

alogy

-1.59

Prev

-1.56

ua

-1.51

irt

-1.48

ulas

-1.48

mine

-1.45

POSITIVE LOGITS

gomery

1.84

 Boone

1.78

Yaz

1.64

utherland

1.55

avorite

1.54

 Dorothy

1.52

 Haram

1.50

 edible

1.49

 enthusi

1.48

phabet

1.47

Activations Density 0.000%

No Known Activations

This feature has no known activations.