INDEX

Explanations

instances of reported speech or quotations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.9.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

0:0.04

1:0.02

2:0.05

3:0.10

4:0.11

5:0.19

6:0.05

7:0.05

8:0.19

9:0.02

10:0.05

11:0.08

Negative Logits

OHN

-2.95

.}

-2.58

 capitalize

-2.38

 Wichita

-2.34

tac

-2.33

rament

-2.32

 comprom

-2.31

.�

-2.27

 redes

-2.25

."[

-2.21

POSITIVE LOGITS

 Whilst

2.88

Labour

2.72

Shape

2.60

Britain

2.53

 Labour

2.52

ITV

2.44

 puberty

2.36

isable

2.33

 Wales

2.30

 Welsh

2.29

Activations Density 0.024%

instances of reported speech or quotations

No Comments

No Known Activations