learning_rate: 5e-05
train_batch_size: 4
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Chrf++	Gen Len
23.7180	1.4706	4	2.4529	10.1334	189.1765
23.7180	2.9412	8	2.4483	9.3038	188.8824
23.7180	4.0	12	2.4491	9.3068	188.7059
23.7180	5.4706	16	2.4468	9.3281	188.9412
23.7180	6.9412	20	2.4453	10.3612	188.7647
23.7180	8.0	24	2.4448	10.1396	188.5882
23.7180	9.4706	28	2.4372	9.6101	189.0588
23.7180	10.9412	32	2.4346	10.4161	188.8824
23.7180	12.0	36	2.4240	11.935	188.7059
23.7180	13.4706	40	2.4080	9.8663	188.5882
23.7180	14.9412	44	2.3919	9.9395	188.4706
23.7180	16.0	48	2.3708	11.9432	188.4706
23.7180	17.4706	52	2.3495	11.677	188.5294
23.7180	18.9412	56	2.3174	10.4155	188.5882
23.7180	20.0	60	2.2911	12.0623	188.5294
23.7180	21.4706	64	2.2371	12.0079	188.4118
23.7180	22.9412	68	2.1985	11.5682	188.5294
23.7180	24.0	72	2.1474	14.7067	188.8824
23.7180	25.4706	76	2.0874	14.6076	188.4706
23.7180	26.9412	80	2.0198	14.5237	188.3529
23.7180	28.0	84	1.9565	17.337	188.2941
23.7180	29.4706	88	1.8893	18.315	188.1176
23.7180	30.9412	92	1.8333	10.8076	187.8235
23.7180	32.0	96	1.7796	9.0608	187.7647
14.9840	33.4706	100	1.7194	14.1862	187.4706
14.9840	34.9412	104	1.6528	13.9242	187.7059
14.9840	36.0	108	1.6024	10.1282	187.7647
14.9840	37.4706	112	1.5534	12.4221	187.5882
14.9840	38.9412	116	1.5073	14.0919	187.5294
14.9840	40.0	120	1.4766	14.3125	187.2353
14.9840	41.4706	124	1.4255	14.7926	186.5294
14.9840	42.9412	128	1.3740	15.0843	183.0588
14.9840	44.0	132	1.3263	16.4486	180.8824
14.9840	45.4706	136	1.2737	16.8679	174.1176
14.9840	46.9412	140	1.2321	17.2518	157.8235
14.9840	48.0	144	1.2062	18.3491	98.7059
14.9840	49.4706	148	1.1740	18.5572	102.7059
14.9840	50.9412	152	1.1186	18.527	112.1176
14.9840	52.0	156	1.0855	20.9775	115.4706
14.9840	53.4706	160	1.0638	18.9739	101.7647
14.9840	54.9412	164	1.0237	17.9036	119.2353
14.9840	56.0	168	0.9828	20.4917	119.1765
14.9840	57.4706	172	0.9463	25.1606	81.2941
14.9840	58.9412	176	0.9221	24.9383	101.8235
14.9840	60.0	180	0.8867	24.8829	94.6471
14.9840	61.4706	184	0.8637	25.6673	98.7647
14.9840	62.9412	188	0.8300	27.4894	76.6471
14.9840	64.0	192	0.7985	28.7274	63.7647
14.9840	65.4706	196	0.7690	30.8172	62.7059
7.3223	66.9412	200	0.7481	33.3092	64.0
7.3223	68.0	204	0.7371	31.6978	70.7647
7.3223	69.4706	208	0.7075	33.8318	54.1765
7.3223	70.9412	212	0.6877	36.3644	58.9412
7.3223	72.0	216	0.6706	35.9372	56.3529
7.3223	73.4706	220	0.6470	35.9255	59.5294
7.3223	74.9412	224	0.6287	36.449	54.3529
7.3223	76.0	228	0.6329	37.1882	63.4706
7.3223	77.4706	232	0.6008	37.5648	68.5882
7.3223	78.9412	236	0.5854	36.3704	56.9412
7.3223	80.0	240	0.5853	40.3231	62.3529
7.3223	81.4706	244	0.5602	42.3926	62.5882
7.3223	82.9412	248	0.5607	42.949	65.2941
7.3223	84.0	252	0.5392	43.7572	64.8235
7.3223	85.4706	256	0.5443	41.7907	65.7647
7.3223	86.9412	260	0.5291	45.8454	63.9412
7.3223	88.0	264	0.5105	44.4988	65.0588
7.3223	89.4706	268	0.5285	44.3357	66.5882
7.3223	90.9412	272	0.5009	44.1722	59.9412
7.3223	92.0	276	0.5026	45.9232	65.1765
7.3223	93.4706	280	0.4770	49.6388	66.2941
7.3223	94.9412	284	0.4962	46.9059	74.0
7.3223	96.0	288	0.4832	47.9596	70.9412
7.3223	97.4706	292	0.5037	49.4734	66.0588
7.3223	98.9412	296	0.4869	49.3019	81.2353
3.4194	100.0	300	0.5080	51.2674	78.1765

Framework versions

Transformers 5.12.1
Pytorch 2.6.0+cu124
Datasets 5.0.0
Tokenizers 0.22.2

Tsedeniya/finetuned_byT5base_pnb

finetuned_byT5base_pnb

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Tsedeniya/
finetuned_byT5base_pnb