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

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.08

1:0.09

2:0.10

3:0.08

4:0.07

5:0.08

6:0.07

7:0.09

8:0.07

9:0.08

10:0.07

11:0.08

Negative Logits

folk

-1.82

odiac

-1.51

 immortality

-1.46

 trem

-1.41

utherford

-1.40

rave

-1.39

DragonMagazine

-1.39

venge

-1.38

Flight

-1.38

Wiki

-1.38

POSITIVE LOGITS

�醒

1.62

 participating

1.54

 Pinterest

1.45

 disadvantaged

1.40

 unsuccessful

1.39

 chall

1.36

 referrals

1.34

 BuzzFeed

1.34

�

1.32

 Supporting

1.30

Activations Density 0.000%

No Known Activations

This feature has no known activations.