INDEX

Explanations

phrases indicating a state of continuity or ongoing action

New Auto-Interp

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

1:0.01

2:0.10

3:0.21

4:0.06

5:0.03

6:0.27

7:0.03

8:0.03

9:0.03

10:0.09

11:0.06

Negative Logits

 ensuring

-1.37

 Cerberus

-1.32

 Schne

-1.32

 shifting

-1.32

 Rosenthal

-1.31

 Sonia

-1.29

 sweeping

-1.26

 reassuring

-1.24

 fingerprints

-1.24

 Scully

-1.23

POSITIVE LOGITS

untled

1.87

istani

1.65

mor

1.60

inance

1.59

yd

1.54

MpServer

1.54

pires

1.53

yss

1.50

姫

1.49

ilo

1.47

Activations Density 0.190%

phrases indicating a state of continuity or ongoing action

No Known Activations