life2lang-small-it

This model is a fine-tuned version of khairi/life2lang-small-it on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0686

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: 0.0012
• train_batch_size: 24
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 192
• 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: cosine
• lr_scheduler_warmup_ratio: 0.2
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss
1.3258	0.0876	100	1.1836
1.316	0.1752	200	1.1828
1.3189	0.2628	300	1.1843
1.3169	0.3504	400	1.1828
1.3264	0.4380	500	1.1839
1.3238	0.5256	600	1.1818
1.3068	0.6132	700	1.1763
1.3194	0.7008	800	1.1752
1.3217	0.7883	900	1.1724
1.3058	0.8759	1000	1.1742
1.3057	0.9635	1100	1.1769
1.2982	1.0508	1200	1.1794
1.3013	1.1384	1300	1.1760
1.2991	1.2260	1400	1.1713
1.3037	1.3136	1500	1.1717
1.2936	1.4012	1600	1.1638
1.291	1.4888	1700	1.1537
1.2765	1.5764	1800	1.1545
1.2865	1.6640	1900	1.1469
1.2692	1.7516	2000	1.1432
1.2734	1.8392	2100	1.1412
1.2703	1.9267	2200	1.1365
1.2517	2.0140	2300	1.1362
1.2447	2.1016	2400	1.1343
1.247	2.1892	2500	1.1260
1.2451	2.2768	2600	1.1310
1.2446	2.3644	2700	1.1269
1.2431	2.4520	2800	1.1149
1.2381	2.5396	2900	1.1142
1.2345	2.6272	3000	1.1165
1.236	2.7148	3100	1.1120
1.2268	2.8024	3200	1.1069
1.2228	2.8900	3300	1.1025
1.2214	2.9776	3400	1.1001
1.212	3.0648	3500	1.0951
1.2121	3.1524	3600	1.0926
1.2111	3.2400	3700	1.0913
1.2153	3.3276	3800	1.0896
1.2094	3.4152	3900	1.0848
1.2035	3.5028	4000	1.0829
1.2056	3.5904	4100	1.0841
1.1978	3.6780	4200	1.0811
1.2037	3.7656	4300	1.0815
1.201	3.8532	4400	1.0773
1.2012	3.9408	4500	1.0754
1.1982	4.0280	4600	1.0761
1.1846	4.1156	4700	1.0733
1.1978	4.2032	4800	1.0714
1.1977	4.2908	4900	1.0712
1.1971	4.3784	5000	1.0707
1.1878	4.4660	5100	1.0695
1.1824	4.5536	5200	1.0692
1.1968	4.6412	5300	1.0688
1.1878	4.7288	5400	1.0686
1.1888	4.8164	5500	1.0685
1.2	4.9040	5600	1.0686
1.1921	4.9916	5700	1.0686

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.4.2
• Tokenizers 0.22.1