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

0:0.07

1:0.06

2:0.07

3:0.08

4:0.08

5:0.09

6:0.08

7:0.07

8:0.10

9:0.09

10:0.07

11:0.08

Negative Logits

quartered

-1.65

quote

-1.55

BuyableInstoreAndOnline

-1.51

Flight

-1.51

llular

-1.50

 Delicious

-1.48

separ

-1.48

cod

-1.48

fixed

-1.46

stable

-1.46

POSITIVE LOGITS

 Voting

1.80

 voting

1.66

 Burnett

1.64

apon

1.63

 franc

1.61

 Ramos

1.60

 Ramirez

1.59

 vote

1.58

 Voter

1.56

 Democracy

1.55

Activations Density 0.000%

No Known Activations

This feature has no known activations.