Revision_Lex_Domain_Meta_XLM_CLS_Data_46k_train

This model is a fine-tuned version of phunganhsang/Revision_Pho_Lexical_46kClsXlm on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3920
• Accuracy: 0.8931
• F1: 0.8848

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: 2e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.0693	100	0.2910	0.8838	0.8751
No log	0.1385	200	0.2848	0.8887	0.8775
No log	0.2078	300	0.2938	0.8807	0.8700
No log	0.2770	400	0.2890	0.8883	0.8790
No log	0.3463	500	0.2815	0.8896	0.8821
No log	0.4155	600	0.2915	0.8850	0.8764
No log	0.4848	700	0.2735	0.8883	0.8784
No log	0.5540	800	0.2709	0.8935	0.8854
No log	0.6233	900	0.2554	0.8944	0.8853
No log	0.6925	1000	0.2848	0.8864	0.8796
No log	0.7618	1100	0.2803	0.8912	0.8848
No log	0.8310	1200	0.2692	0.8960	0.8867
No log	0.9003	1300	0.2844	0.8909	0.8840
No log	0.9695	1400	0.3017	0.8850	0.8781
0.1981	1.0388	1500	0.3135	0.8845	0.8777
0.1981	1.1080	1600	0.3110	0.8845	0.8709
0.1981	1.1773	1700	0.2880	0.8897	0.8821
0.1981	1.2465	1800	0.3068	0.8790	0.8718
0.1981	1.3158	1900	0.3204	0.8865	0.8797
0.1981	1.3850	2000	0.3047	0.8844	0.8772
0.1981	1.4543	2100	0.2865	0.8881	0.8812
0.1981	1.5235	2200	0.2929	0.8937	0.8853
0.1981	1.5928	2300	0.2757	0.8986	0.8896
0.1981	1.6620	2400	0.2846	0.8965	0.8882
0.1981	1.7313	2500	0.3028	0.8941	0.8855
0.1981	1.8006	2600	0.2814	0.8973	0.8886
0.1981	1.8698	2700	0.3320	0.8819	0.8756
0.1981	1.9391	2800	0.2952	0.8955	0.8876
0.1472	2.0083	2900	0.3051	0.8937	0.8862
0.1472	2.0776	3000	0.3259	0.8882	0.8802
0.1472	2.1468	3100	0.3346	0.8917	0.8841
0.1472	2.2161	3200	0.3425	0.8898	0.8819
0.1472	2.2853	3300	0.3223	0.8957	0.8864
0.1472	2.3546	3400	0.3413	0.8877	0.8794
0.1472	2.4238	3500	0.3199	0.8943	0.8858
0.1472	2.4931	3600	0.3416	0.8897	0.8812
0.1472	2.5623	3700	0.3210	0.8914	0.8835
0.1472	2.6316	3800	0.3310	0.8940	0.8864
0.1472	2.7008	3900	0.3175	0.8913	0.8833
0.1472	2.7701	4000	0.3321	0.8907	0.8832
0.1472	2.8393	4100	0.3256	0.8912	0.8816
0.1472	2.9086	4200	0.3399	0.8929	0.8844
0.1472	2.9778	4300	0.3437	0.8861	0.8785
0.1142	3.0471	4400	0.3319	0.8956	0.8876
0.1142	3.1163	4500	0.3660	0.8904	0.8824
0.1142	3.1856	4600	0.3800	0.8895	0.8806
0.1142	3.2548	4700	0.3713	0.8883	0.8801
0.1142	3.3241	4800	0.3822	0.8904	0.8824
0.1142	3.3934	4900	0.3707	0.8891	0.8817
0.1142	3.4626	5000	0.3430	0.8948	0.8860
0.1142	3.5319	5100	0.3499	0.8937	0.8852
0.1142	3.6011	5200	0.3680	0.8908	0.8831
0.1142	3.6704	5300	0.3659	0.8917	0.8835
0.1142	3.7396	5400	0.3742	0.8912	0.8826
0.1142	3.8089	5500	0.3677	0.8926	0.8847
0.1142	3.8781	5600	0.3703	0.8923	0.8838
0.1142	3.9474	5700	0.3755	0.8913	0.8837
0.0892	4.0166	5800	0.3628	0.8927	0.8851
0.0892	4.0859	5900	0.3784	0.8929	0.8853
0.0892	4.1551	6000	0.3820	0.8906	0.8825
0.0892	4.2244	6100	0.3857	0.8932	0.8850
0.0892	4.2936	6200	0.3911	0.8930	0.8847
0.0892	4.3629	6300	0.3869	0.8929	0.8845
0.0892	4.4321	6400	0.4038	0.8896	0.8823
0.0892	4.5014	6500	0.4114	0.8900	0.8816
0.0892	4.5706	6600	0.4088	0.8908	0.8829
0.0892	4.6399	6700	0.4003	0.8921	0.8831
0.0892	4.7091	6800	0.4021	0.8929	0.8845
0.0892	4.7784	6900	0.3962	0.8922	0.8841
0.0892	4.8476	7000	0.3918	0.8929	0.8843
0.0892	4.9169	7100	0.3921	0.8937	0.8855
0.0892	4.9861	7200	0.3920	0.8931	0.8848

Framework versions

• Transformers 5.3.0
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.2