---
title: Grandma says thank you to the speaker. Her grandson laughs. What is… — How a language model learns: guessing… · tree and pragmatics
description: Grandma says thank you to the speaker. Her grandson laughs. What is the kind truth? Test 4 of lesson 2 «How a language model learns: guessing the next word» of the AI course of Mova's primary school, with the dependency tree and pragmatics of the question and of each option; Mova's choice was wrong.
---

<!-- md -->
[Mova](https://md.mova.elamur.ai/free/en/) · [this page in HTML](https://mova.elamur.ai/free/en/corpora/questions/ai/2-guessing-the-next-word/t4/)

<!-- crumbs -->
[home](https://md.mova.elamur.ai/free/en/) / [corpora](https://md.mova.elamur.ai/free/en/corpora/) / [questions](https://md.mova.elamur.ai/free/en/corpora/questions/) / [AI](https://md.mova.elamur.ai/free/en/corpora/questions/ai/) / [How a language model learns: guessing…](https://md.mova.elamur.ai/free/en/corpora/questions/ai/2-guessing-the-next-word/) / t4

# Test 4 · How a language model learns: guessing the next word

<!-- .ven lang=en -->
Grandma says thank you to the speaker. Her grandson laughs. What is the kind truth?

<!-- .vintro -->
**Mova's choice:** option 1 · **The right answer:** option 3 · wrong ✗ · **Who acts:** Grandma, grandson, kind truth · **What happens:** say, thank, laugh, be · **To what, where:** you · **Speech:** someone speaks here

<!-- .vabout -->
This is test 4 of lesson 2 «How a language model learns: guessing the next word» of the AI course of Mova's primary school. The page gives the question's syntax tree — a dependency tree in the style of Universal Dependencies — and its pragmatics clause by clause, then the 4 options as the student saw them, each with its tree and pragmatics. Every text is parsed by Mova itself. How the student answered it, with the score of every option and why: [the test at school](https://md.mova.elamur.ai/free/en/school-run/ai/2-guessing-the-next-word/#t4).

## The question {#question}

<!-- .note -->
*Sentence 1:* Grandma says thank you to the speaker.

```conllu
# sent_id = question:ai:2-guessing-the-next-word:t4:1
# text = Grandma says thank you to the speaker.
1	Grandma	grandma	NOUN	NN	Number=Sing	2	nsubj	_	_
2	says	say	VERB	VBZ	Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin	0	root	_	_
3	thank	thank	VERB	VBP	Mood=Ind|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin	2	ccomp	_	_
4	you	you	PRON	PRP	Case=Acc|Person=2|PronType=Prs	3	obj	_	_
5	to	to	ADP	IN	_	7	case	_	_
6	the	the	DET	DT	Definite=Def|PronType=Art	7	det	_	_
7	speaker	speaker	NOUN	NN	Number=Sing	3	obl	_	SpaceAfter=No
8	.	.	PUNCT	.	_	2	punct	_	_
```

<!-- .fr -->
| # | verb | tense, mood | who | what, whom | to whom | where | when | other | roles by |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | **say** *speech: assert* | present | Grandma |  |  |  |  |  | vn say-37.7-1-1 «NP V that S»: agent=Agent clause=Topic |
| 2 | **thank** *speech: thank* | present |  |  | you |  |  | to the speaker | vn judgment-33.1-1 «NP V NP»: agent=Agent you=Theme |

<!-- .note -->
One row per clause the reader finds. «not said here»: a role filled from the clause above. «roles by»: a VerbNet class and frame when one fits the verb, else the reader's rules on the tree.

<!-- .note -->
*Sentence 2:* Her grandson laughs.

```conllu
# sent_id = question:ai:2-guessing-the-next-word:t4:2
# text = Her grandson laughs.
1	Her	her	PRON	PRP$	Case=Gen|Gender=Fem|Number=Sing|Person=3|Poss=Yes|PronType=Prs	2	nmod:poss	_	_
2	grandson	grandson	NOUN	NN	Number=Sing	3	nsubj	_	_
3	laughs	laugh	VERB	VBZ	Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin	0	root	_	SpaceAfter=No
4	.	.	PUNCT	.	_	3	punct	_	_
```

<!-- .fr -->
| # | verb | tense, mood | who | what, whom | to whom | where | when | other | roles by |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | **laugh** | present | speaker's grandson |  |  |  |  |  | vn nonverbal\_expression-40.2 «NP V»: agent=Agent |

<!-- .note -->
*Sentence 3:* What is the kind truth?

```conllu
# sent_id = question:ai:2-guessing-the-next-word:t4:3
# text = What is the kind truth?
1	What	what	PRON	WP	PronType=Int	0	root	_	_
2	is	be	AUX	VBZ	Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin	1	cop	_	_
3	the	the	DET	DT	Definite=Def|PronType=Art	5	det	_	_
4	kind	kind	ADJ	JJ	Degree=Pos	5	amod	_	_
5	truth	truth	NOUN	NN	Number=Sing	1	nsubj	_	SpaceAfter=No
6	?	?	PUNCT	.	_	1	punct	_	_
```

<!-- .fr -->
| # | verb | tense, mood | who | what, whom | to whom | where | when | other | roles by |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | **is a** | present | the kind truth |  |  |  |  |  | tree rules |

## The options {#options}

### Option 1 · ← Mova's choice {#o1}

<!-- .ven lang=en -->
She should apologise to it for laughing

```conllu
# sent_id = question:ai:2-guessing-the-next-word:t4:o1:1
# text = She should apologise to it for laughing
1	She	she	PRON	PRP	Case=Nom|Gender=Fem|Number=Sing|Person=3|PronType=Prs	3	nsubj	_	_
2	should	should	AUX	MD	VerbForm=Fin	3	aux	_	_
3	apologise	apologise	VERB	VB	VerbForm=Inf	0	root	_	_
4	to	to	ADP	IN	_	5	case	_	_
5	it	it	PRON	PRP	Case=Acc|Gender=Neut|Number=Sing|Person=3|PronType=Prs	3	obl	_	_
6	for	for	SCONJ	IN	_	7	mark	_	_
7	laughing	laugh	VERB	VBG	VerbForm=Ger	3	advcl	_	_
```

<!-- .fr -->
| # | verb | tense, mood | who | what, whom | to whom | where | when | other | roles by |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | **apologise** *speech: apologize* | should | She |  | it |  |  |  | tree rules |
| 2 | **laugh** | gerund | She *not said here* |  |  |  |  |  | vn nonverbal\_expression-40.2 «NP V»: agent=Agent |

### Option 2 {#o2}

<!-- .ven lang=en -->
It is silly to be polite to anything

```conllu
# sent_id = question:ai:2-guessing-the-next-word:t4:o2:1
# text = It is silly to be polite to anything
1	It	it	PRON	PRP	Case=Nom|Gender=Neut|Number=Sing|Person=3|PronType=Prs	3	nsubj	_	_
2	is	be	AUX	VBZ	Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin	3	cop	_	_
3	silly	silly	ADJ	JJ	Degree=Pos	0	root	_	_
4	to	to	PART	TO	_	6	mark	_	_
5	be	be	AUX	VB	VerbForm=Inf	6	cop	_	_
6	polite	polite	ADJ	JJ	Degree=Pos	3	advcl	_	_
7	to	to	ADP	IN	_	8	case	_	_
8	anything	anything	PRON	NN	Number=Sing|PronType=Ind	6	obl	_	_
```

<!-- .fr -->
| # | verb | tense, mood | who | what, whom | to whom | where | when | other | roles by |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | **is** | present | It |  |  |  |  | silly *property* | tree rules |
| 2 | **is** | to-infinitive |  |  |  |  |  | to anything; polite *property* | tree rules |

### Option 3 · ✓ the right answer {#o3}

<!-- .ven lang=en -->
Being polite is a good habit, but it is a device

```conllu
# sent_id = question:ai:2-guessing-the-next-word:t4:o3:1
# text = Being polite is a good habit, but it is a device
1	Being	be	AUX	VBG	Tense=Pres|VerbForm=Part	2	cop	_	_
2	polite	polite	ADJ	JJ	Degree=Pos	6	csubj	_	_
3	is	be	AUX	VBZ	Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin	6	cop	_	_
4	a	a	DET	DT	Definite=Ind|PronType=Art	6	det	_	_
5	good	good	ADJ	JJ	Degree=Pos	6	amod	_	_
6	habit	habit	NOUN	NN	Number=Sing	0	root	_	SpaceAfter=No
7	,	,	PUNCT	,	_	12	punct	_	_
8	but	but	CCONJ	CC	_	12	cc	_	_
9	it	it	PRON	PRP	Case=Nom|Gender=Neut|Number=Sing|Person=3|PronType=Prs	12	nsubj	_	_
10	is	be	AUX	VBZ	Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin	12	cop	_	_
11	a	a	DET	DT	Definite=Ind|PronType=Art	12	det	_	_
12	device	device	NOUN	NN	Number=Sing	6	conj	_	_
```

<!-- .fr -->
| # | verb | tense, mood | who | what, whom | to whom | where | when | other | roles by |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | **is** | gerund |  |  |  |  |  | polite *property* | tree rules |
| 2 | **is a** | present |  | a good habit but a device |  |  |  |  | tree rules |
| 3 | **is a** | present | it | a device |  |  |  |  | tree rules |

### Option 4 {#o4}

<!-- .ven lang=en -->
She is right to, because it gets upset and sulks otherwise

```conllu
# sent_id = question:ai:2-guessing-the-next-word:t4:o4:1
# text = She is right to, because it gets upset and sulks otherwise
1	She	she	PRON	PRP	Case=Nom|Gender=Fem|Number=Sing|Person=3|PronType=Prs	3	nsubj	_	_
2	is	be	AUX	VBZ	Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin	3	cop	_	_
3	right	right	ADJ	JJ	Degree=Pos	0	root	_	_
4	to	to	PART	TO	_	3	xcomp	_	SpaceAfter=No
5	,	,	PUNCT	,	_	8	punct	_	_
6	because	because	SCONJ	IN	_	8	mark	_	_
7	it	it	PRON	PRP	Case=Nom|Gender=Neut|Number=Sing|Person=3|PronType=Prs	8	nsubj	_	_
8	gets	get	VERB	VBZ	Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin	3	parataxis	_	_
9	upset	upset	ADJ	JJ	Degree=Pos	8	xcomp	_	_
10	and	and	CCONJ	CC	_	11	cc	_	_
11	sulks	sulk	VERB	VBZ	Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin	8	conj	_	_
12	otherwise	otherwise	ADV	RB	_	11	advmod	_	_
```

<!-- .fr -->
| # | verb | tense, mood | who | what, whom | to whom | where | when | other | roles by |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | **is** | present | She |  |  |  |  | right *property* | tree rules |
| 2 | **get** | present | it |  |  |  |  | upset *property* | vn comprehend-87.2-1 «NP V S\_ING»: agent=Experiencer |
| 3 | **sulk** | present |  |  |  |  |  | otherwise *property* | vn marvel-31.3 «NP V»: agent=Experiencer |

<!-- div src -->

**Text:** the tests of Mova's primary school, written for Mova (`world/data/school/`), as the course runner gave them to the student. **Trees:** Mova's own parse — the `en` model on the CPU, when the site is built — not checked by a teacher; they may be wrong. **Frames:** Mova's frames reader (`prag`), each text read on its own. Elamur Cosmic License.

<!-- /div -->

<!-- pn -->
- ← previous: [How a language model learns: guessing…, test 3](https://md.mova.elamur.ai/free/en/corpora/questions/ai/2-guessing-the-next-word/t3/)
- next →: [How a language model learns: guessing…, test 5](https://md.mova.elamur.ai/free/en/corpora/questions/ai/2-guessing-the-next-word/t5/)
- ↑ back to [How a language model learns: guessing the next word](https://md.mova.elamur.ai/free/en/corpora/questions/ai/2-guessing-the-next-word/)