You can deploy this model and start earning money today!

contemmcm/
676180cb9b08ef929ecc2d7af3d50c14

generated_from_trainer

Views

No views yet

676180cb9b08ef929ecc2d7af3d50c14

This model is a fine-tuned version of google/mt5-small on the Helsinki-NLP/opus_books [en-ru] dataset. It achieves the following results on the evaluation set:

Loss: 1.8945
Data Size: 1.0
Epoch Runtime: 66.6586
Bleu: 6.8545

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 4
total_train_batch_size: 32
total_eval_batch_size: 32
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: constant
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Bleu
No log	0	0	27.6965	0	5.8895	0.0035
No log	1	437	26.8773	0.0078	6.3942	0.0032
No log	2	874	23.1427	0.0156	7.5209	0.0080
No log	3	1311	17.0758	0.0312	8.5876	0.0056
No log	4	1748	11.1508	0.0625	10.4148	0.0066
12.128	5	2185	6.8811	0.125	14.1073	0.0295
7.8068	6	2622	4.0391	0.25	20.9845	0.0551
4.442	7	3059	3.0637	0.5	35.4912	1.3072
3.7338	8.0	3496	2.7166	1.0	65.4882	2.2786
3.4468	9.0	3933	2.5689	1.0	64.4147	2.8379
3.2427	10.0	4370	2.4780	1.0	65.4351	3.1619
3.1548	11.0	4807	2.4012	1.0	64.2597	3.4374
3.0509	12.0	5244	2.3484	1.0	65.8841	3.6792
2.9277	13.0	5681	2.3068	1.0	64.8017	3.8500
2.8542	14.0	6118	2.2708	1.0	65.4200	4.0126
2.7918	15.0	6555	2.2341	1.0	65.2679	4.1396
2.7146	16.0	6992	2.2035	1.0	66.7773	4.3280
2.7101	17.0	7429	2.1841	1.0	65.2585	4.4390
2.6344	18.0	7866	2.1594	1.0	65.5984	4.6303
2.568	19.0	8303	2.1393	1.0	67.1122	4.8554
2.5458	20.0	8740	2.1222	1.0	66.5826	5.0379
2.495	21.0	9177	2.1019	1.0	68.3891	5.2442
2.4473	22.0	9614	2.0850	1.0	66.2306	5.3179
2.4353	23.0	10051	2.0742	1.0	65.3412	5.4400
2.3907	24.0	10488	2.0585	1.0	67.7938	5.4989
2.4033	25.0	10925	2.0454	1.0	64.7211	5.6262
2.3446	26.0	11362	2.0347	1.0	64.5664	5.7102
2.3442	27.0	11799	2.0264	1.0	64.5435	5.8278
2.2789	28.0	12236	2.0170	1.0	65.0696	5.9744
2.2483	29.0	12673	2.0040	1.0	65.3662	5.9919
2.2154	30.0	13110	2.0004	1.0	64.7863	5.9901
2.2096	31.0	13547	1.9864	1.0	66.0260	6.1058
2.2434	32.0	13984	1.9837	1.0	64.4548	6.1465
2.1895	33.0	14421	1.9713	1.0	65.4305	6.2313
2.1619	34.0	14858	1.9642	1.0	68.3454	6.2970
2.1225	35.0	15295	1.9595	1.0	71.6646	6.3330
2.0963	36.0	15732	1.9528	1.0	70.7175	6.3455
2.0945	37.0	16169	1.9495	1.0	69.7594	6.4101
2.0548	38.0	16606	1.9390	1.0	68.4443	6.4673
2.0481	39.0	17043	1.9331	1.0	70.9399	6.4760
2.0254	40.0	17480	1.9293	1.0	69.1638	6.5702
2.0301	41.0	17917	1.9316	1.0	69.5852	6.5667
1.9951	42.0	18354	1.9223	1.0	69.9902	6.5834
1.9953	43.0	18791	1.9195	1.0	71.0596	6.6935
1.9566	44.0	19228	1.9091	1.0	69.5052	6.7176
1.9871	45.0	19665	1.9095	1.0	70.5844	6.7239
1.9231	46.0	20102	1.9049	1.0	64.4600	6.7174
1.9222	47.0	20539	1.8987	1.0	65.1935	6.7957
1.9208	48.0	20976	1.8974	1.0	66.3609	6.8146
1.8593	49.0	21413	1.8931	1.0	65.1668	6.8129
1.8743	50.0	21850	1.8945	1.0	66.6586	6.8545

Framework versions

Transformers 4.57.0
Pytorch 2.8.0+cu128
Datasets 4.2.0
Tokenizers 0.22.1

contemmcm/676180cb9b08ef929ecc2d7af3d50c14

676180cb9b08ef929ecc2d7af3d50c14

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

contemmcm/
676180cb9b08ef929ecc2d7af3d50c14