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

1:0.05

2:0.09

3:0.08

4:0.08

5:0.08

6:0.08

7:0.08

8:0.08

9:0.09

10:0.08

11:0.08

Negative Logits

 upbeat

-1.62

Attorney

-1.58

OFF

-1.51

cand

-1.49

Secretary

-1.47

options

-1.47

facebook

-1.45

earchers

-1.44

staff

-1.43

Privacy

-1.43

POSITIVE LOGITS

 Plague

1.60

 Lazarus

1.57

 Pinball

1.52

gans

1.52

Va

1.50

 Invention

1.47

 Renaissance

1.47

 empires

1.46

sem

1.44

 Nether

1.43

Activations Density 0.000%

No Known Activations

This feature has no known activations.