INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.09

1:0.07

2:0.10

3:0.08

4:0.07

5:0.06

6:0.08

7:0.06

8:0.09

9:0.08

10:0.08

11:0.08

Negative Logits

uses

-1.38

 prest

-1.28

tre

-1.25

google

-1.24

ille

-1.23

agra

-1.22

 thereof

-1.22

insert

-1.21

 Draft

-1.20

config

-1.20

POSITIVE LOGITS

ollar

1.65

theless

1.56

 underestimate

1.45

ONSORED

1.39

 mentality

1.38

 Duterte

1.37

ccording

1.31

 underestimated

1.29

 Krish

1.28

 animosity

1.27

Activations Density 0.000%

No Known Activations

This feature has no known activations.