INDEX

Explanations

phrases indicating uncertainty or indecision, especially related to voter opinions

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_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.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

1:0.15

2:0.06

3:0.08

4:0.08

5:0.02

6:0.14

7:0.19

8:0.04

9:0.04

10:0.05

11:0.08

Negative Logits

 GOODMAN

-2.72

Lex

-2.60

tnc

-2.35

ARA

-2.35

gression

-2.30

assetsadobe

-2.22

OPS

-2.15

Tup

-2.13

 Quadro

-2.12

 examinations

-2.10

POSITIVE LOGITS

 undecided

4.59

ecided

2.73

 leaning

2.60

kell

2.45

HRC

2.37

Veg

2.33

vote

2.33

 scept

2.33

 withd

2.27

 spoiler

2.24

Activations Density 0.001%

phrases indicating uncertainty or indecision, especially related to voter opinions

No Comments

No Known Activations