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

0:0.07

1:0.08

2:0.08

3:0.07

4:0.08

5:0.07

6:0.08

7:0.08

8:0.08

9:0.08

10:0.08

11:0.09

Negative Logits

accompanied

-1.39

Guest

-1.37

lear

-1.36

requisite

-1.31

 Messi

-1.31

 Xander

-1.31

Auth

-1.29

achus

-1.28

tu

-1.26

notations

-1.25

POSITIVE LOGITS

 sqor

1.56

ylon

1.51

alach

1.48

 commercially

1.36

ombo

1.33

imental

1.32

ournals

1.32

 cass

1.31

 collaborations

1.29

ripp

1.27

Activations Density 0.000%

No Known Activations

This feature has no known activations.