INDEX

Explanations

phrases indicating attentiveness or awareness

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

Head Attr Weights

0:0.01

1:0.02

2:0.05

3:0.06

4:0.20

5:0.04

6:0.04

7:0.31

8:0.05

9:0.05

10:0.06

11:0.05

Negative Logits

uct

-1.52

Plex

-1.51

attach

-1.45

uming

-1.45

installed

-1.42

 financed

-1.42

 ware

-1.41

apsed

-1.41

disk

-1.40

acent

-1.39

POSITIVE LOGITS

 footsteps

1.83

 caution

1.78

 chants

1.73

 cues

1.73

 lessons

1.71

 cautiously

1.68

龍契士

1.58

 warnings

1.58

 wisely

1.52

 directives

1.49

Activations Density 0.000%

phrases indicating attentiveness or awareness

No Known Activations