INDEX

Explanations

phrases related to inconsistencies or contradictions

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

0:0.02

1:0.50

2:0.06

3:0.04

4:0.05

5:0.04

6:0.03

7:0.05

8:0.02

9:0.05

10:0.05

11:0.05

Negative Logits

artney

-2.90

 toget

-2.36

iliar

-2.34

donald

-2.34

ipes

-2.31

 liking

-2.28

 encount

-2.27

lear

-2.26

usually

-2.19

 includ

-2.14

POSITIVE LOGITS

no

6.66

no

6.21

NO

6.02

No

5.84

No

5.83

NO

5.34

nos

3.95

YES

3.85

Nos

3.78

 none

3.70

Activations Density 0.105%

phrases related to inconsistencies or contradictions

No Comments

No Known Activations