You can deploy this model and start earning money today!

contemmcm/
17d50d1bdf9cc52c4ce2c12eeb42721b

generated_from_trainer

Views

No views yet

17d50d1bdf9cc52c4ce2c12eeb42721b

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

Loss: 2.5114
Data Size: 1.0
Epoch Runtime: 14.6190
Bleu: 2.0755

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	27.2708	0	1.8840	0.0048
No log	1	88	26.7347	0.0078	2.1187	0.0053
No log	2	176	25.0164	0.0156	2.8001	0.0047
No log	3	264	23.0135	0.0312	3.4500	0.0083
No log	4	352	19.6700	0.0625	3.9502	0.0063
No log	5	440	15.9559	0.125	4.7035	0.0079
1.6812	6	528	12.2228	0.25	7.0401	0.0073
5.2843	7	616	8.5098	0.5	9.9755	0.0124
7.811	8.0	704	4.9263	1.0	16.4943	0.0108
6.1827	9.0	792	3.6017	1.0	14.1179	0.1010
4.71	10.0	880	3.1982	1.0	14.2631	0.2850
4.2434	11.0	968	3.0572	1.0	14.5411	0.5954
3.994	12.0	1056	2.9829	1.0	14.6358	0.6089
3.9015	13.0	1144	2.9188	1.0	14.9247	0.6874
3.7647	14.0	1232	2.8737	1.0	15.2077	0.7936
3.6643	15.0	1320	2.8368	1.0	13.8218	0.9224
3.568	16.0	1408	2.8072	1.0	13.8018	0.9896
3.5063	17.0	1496	2.7854	1.0	14.6322	0.9964
3.4393	18.0	1584	2.7576	1.0	14.1085	1.1078
3.4149	19.0	1672	2.7373	1.0	13.9114	1.1057
3.3547	20.0	1760	2.7191	1.0	14.0701	1.1366
3.3051	21.0	1848	2.7025	1.0	13.8557	1.2252
3.2991	22.0	1936	2.6882	1.0	14.3848	1.3063
3.2379	23.0	2024	2.6744	1.0	13.7411	1.3361
3.2056	24.0	2112	2.6632	1.0	14.1662	1.3492
3.1705	25.0	2200	2.6547	1.0	14.5135	1.4291
3.1403	26.0	2288	2.6403	1.0	14.8698	1.5161
3.0918	27.0	2376	2.6279	1.0	14.5250	1.5594
3.042	28.0	2464	2.6161	1.0	14.6435	1.5715
3.0391	29.0	2552	2.6059	1.0	14.9035	1.5989
2.9916	30.0	2640	2.6035	1.0	13.5504	1.6349
2.9836	31.0	2728	2.5953	1.0	13.9975	1.6877
2.9478	32.0	2816	2.5868	1.0	14.0837	1.7361
2.9295	33.0	2904	2.5818	1.0	14.3945	1.7080
2.9084	34.0	2992	2.5766	1.0	14.2678	1.7475
2.8892	35.0	3080	2.5707	1.0	13.9927	1.7799
2.8425	36.0	3168	2.5613	1.0	14.6779	1.8156
2.8465	37.0	3256	2.5587	1.0	14.7077	1.8651
2.8295	38.0	3344	2.5555	1.0	13.8400	1.8699
2.7896	39.0	3432	2.5530	1.0	14.2273	1.8924
2.7825	40.0	3520	2.5451	1.0	14.1114	1.8544
2.7608	41.0	3608	2.5432	1.0	14.3696	1.8477
2.7407	42.0	3696	2.5390	1.0	14.6226	1.9139
2.7036	43.0	3784	2.5292	1.0	15.1214	1.9218
2.6947	44.0	3872	2.5309	1.0	14.9670	1.9204
2.67	45.0	3960	2.5258	1.0	15.8521	2.0065
2.6762	46.0	4048	2.5240	1.0	13.5922	1.9647
2.6356	47.0	4136	2.5190	1.0	14.3999	2.0290
2.6297	48.0	4224	2.5176	1.0	14.2129	2.0652
2.6242	49.0	4312	2.5155	1.0	14.3901	2.0852
2.6246	50.0	4400	2.5114	1.0	14.6190	2.0755

Framework versions

Transformers 4.57.0
Pytorch 2.8.0+cu128
Datasets 4.2.0
Tokenizers 0.22.1

contemmcm/17d50d1bdf9cc52c4ce2c12eeb42721b

17d50d1bdf9cc52c4ce2c12eeb42721b

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

contemmcm/
17d50d1bdf9cc52c4ce2c12eeb42721b