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

0:0.08

1:0.07

2:0.08

3:0.08

4:0.07

5:0.08

6:0.07

7:0.09

8:0.08

9:0.08

10:0.08

11:0.08

Negative Logits

 adjourn

-1.66

 setback

-1.47

 deserted

-1.44

 manoeuv

-1.41

 slope

-1.39

 erection

-1.38

 Flavoring

-1.36

 breached

-1.35

 trailer

-1.35

 burnt

-1.35

POSITIVE LOGITS

instead

1.70

ال

1.54

agine

1.53

ALS

1.51

ガ

1.50

codes

1.47

 Greenberg

1.42

 Rodrig

1.42

FOX

1.40

DOC

1.40

Activations Density 0.000%

No Known Activations

This feature has no known activations.