v3_articles_single_base

This model is a fine-tuned version of xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.6527
• Accuracy: 0.4034
• F1: 0.4216

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: 128
• eval_batch_size: 128
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
9.9788	0.1909	500	9.9691	0.0000	0.0000
9.8491	0.3818	1000	9.8108	0.0064	0.0001
9.559	0.5727	1500	9.5123	0.0064	0.0001
9.2051	0.7637	2000	9.1544	0.0064	0.0002
8.9159	0.9546	2500	8.8713	0.0064	0.0002
8.6687	1.1455	3000	8.6692	0.0124	0.0014
8.5236	1.3364	3500	8.5122	0.0205	0.0027
8.3933	1.5273	4000	8.3659	0.0243	0.0048
8.2382	1.7182	4500	8.1938	0.0290	0.0062
8.0455	1.9091	5000	7.9894	0.0427	0.0130
7.6723	2.1000	5500	7.7463	0.0576	0.0180
7.4806	2.2910	6000	7.4716	0.0756	0.0288
7.195	2.4819	6500	7.1386	0.1029	0.0444
6.9325	2.6728	7000	6.7902	0.1262	0.0611
6.4846	2.8637	7500	6.4323	0.1423	0.0724
6.0024	3.0546	8000	6.0699	0.1601	0.0882
5.7705	3.2455	8500	5.7652	0.1719	0.0966
5.453	3.4364	9000	5.4848	0.1852	0.1095
5.24	3.6273	9500	5.2315	0.1997	0.1260
5.0462	3.8183	10000	5.0269	0.2122	0.1342
4.8026	4.0092	10500	4.8465	0.2201	0.1456
4.525	4.2001	11000	4.6870	0.2301	0.1564
4.4423	4.3910	11500	4.5442	0.2438	0.1693
4.2558	4.5819	12000	4.4112	0.2503	0.1779
4.138	4.7728	12500	4.2965	0.2606	0.1874
4.094	4.9637	13000	4.1810	0.2689	0.1987
3.8333	5.1546	13500	4.0969	0.2746	0.2057
3.8822	5.3456	14000	4.0132	0.2790	0.2119
3.7825	5.5365	14500	3.9188	0.2896	0.2238
3.5839	5.7274	15000	3.8476	0.2955	0.2290
3.5884	5.9183	15500	3.7643	0.3035	0.2426
3.4725	6.1092	16000	3.6964	0.3109	0.2506
3.4295	6.3001	16500	3.6443	0.3138	0.2566
3.3089	6.4910	17000	3.5774	0.3209	0.2662
3.3234	6.6819	17500	3.5313	0.3238	0.2669
3.2996	6.8729	18000	3.4774	0.3304	0.2778
3.1088	7.0638	18500	3.4367	0.3317	0.2794
3.049	7.2547	19000	3.3922	0.3371	0.2885
3.0197	7.4456	19500	3.3527	0.3384	0.2912
3.0054	7.6365	20000	3.3090	0.3444	0.2966
2.9578	7.8274	20500	3.2725	0.3493	0.3054
2.9004	8.0183	21000	3.2319	0.3531	0.3111
2.79	8.2092	21500	3.1944	0.3591	0.3220
2.7671	8.4002	22000	3.1643	0.3583	0.3233
2.7441	8.5911	22500	3.1439	0.3617	0.3287
2.7601	8.7820	23000	3.1184	0.3658	0.3289
2.7741	8.9729	23500	3.0845	0.3651	0.3360
2.7503	9.1638	24000	3.0580	0.3686	0.3471
2.603	9.3547	24500	3.0327	0.3706	0.3438
2.5722	9.5456	25000	3.0078	0.3726	0.3496
2.5372	9.7365	25500	2.9964	0.3757	0.3482
2.5411	9.9275	26000	2.9608	0.3737	0.3572
2.4594	10.1184	26500	2.9482	0.3752	0.3590
2.4336	10.3093	27000	2.9390	0.3783	0.3601
2.4163	10.5002	27500	2.9005	0.3834	0.3709
2.4297	10.6911	28000	2.8966	0.3837	0.3658
2.4118	10.8820	28500	2.8715	0.3843	0.3743
2.2559	11.0729	29000	2.8604	0.3813	0.3726
2.2856	11.2638	29500	2.8406	0.3872	0.3818
2.3113	11.4548	30000	2.8189	0.3866	0.3846
2.311	11.6457	30500	2.8048	0.3911	0.3880
2.2357	11.8366	31000	2.7862	0.3907	0.3912
2.1633	12.0275	31500	2.7756	0.3905	0.3929
2.1658	12.2184	32000	2.7612	0.3917	0.3951
2.1555	12.4093	32500	2.7593	0.3950	0.3978
2.1318	12.6002	33000	2.7390	0.3966	0.4021
2.1729	12.7911	33500	2.7257	0.3940	0.4021
2.112	12.9821	34000	2.7123	0.3956	0.4059
2.1104	13.1730	34500	2.7084	0.3956	0.4077
2.059	13.3639	35000	2.6962	0.3999	0.4132
2.0413	13.5548	35500	2.6910	0.3986	0.4100
2.0357	13.7457	36000	2.6787	0.3990	0.4110
2.0525	13.9366	36500	2.6616	0.4019	0.4180
1.918	14.1275	37000	2.6545	0.3998	0.4161
1.9494	14.3184	37500	2.6527	0.4034	0.4216
1.9953	14.5094	38000	2.6409	0.4029	0.4233
1.9497	14.7003	38500	2.6393	0.4069	0.4243
1.9438	14.8912	39000	2.6196	0.4040	0.4273
1.8923	15.0821	39500	2.6127	0.4074	0.4344
1.8606	15.2730	40000	2.6162	0.4087	0.4313
1.9162	15.4639	40500	2.6046	0.4053	0.4326
1.8617	15.6548	41000	2.6003	0.4089	0.4348
1.8639	15.8457	41500	2.5879	0.4111	0.4379
1.7972	16.0367	42000	2.5834	0.4083	0.4392
1.762	16.2276	42500	2.5844	0.4085	0.4381
1.777	16.4185	43000	2.5691	0.4092	0.4433
1.8193	16.6094	43500	2.5720	0.4094	0.4437
1.7783	16.8003	44000	2.5529	0.4128	0.4484
1.7733	16.9912	44500	2.5468	0.4104	0.4490
1.7001	17.1821	45000	2.5509	0.4093	0.4487
1.7282	17.3730	45500	2.5500	0.4132	0.4497
1.7175	17.5640	46000	2.5405	0.4104	0.4498
1.7631	17.7549	46500	2.5405	0.4127	0.4498
1.6979	17.9458	47000	2.5342	0.4105	0.4513
1.6255	18.1367	47500	2.5347	0.4148	0.4526
1.65	18.3276	48000	2.5238	0.4126	0.4536
1.6412	18.5185	48500	2.5239	0.4143	0.4565
1.6252	18.7094	49000	2.5211	0.4151	0.4554
1.6629	18.9003	49500	2.5078	0.4160	0.4607
1.5831	19.0913	50000	2.5184	0.4143	0.4570
1.5809	19.2822	50500	2.5014	0.4155	0.4616
1.5816	19.4731	51000	2.5012	0.4176	0.4634
1.5276	19.6640	51500	2.5002	0.4180	0.4631
1.582	19.8549	52000	2.4901	0.4161	0.4627
1.5468	20.0458	52500	2.4929	0.4180	0.4669
1.5483	20.2367	53000	2.4910	0.4183	0.4667
1.5168	20.4276	53500	2.4932	0.4161	0.4656

Framework versions

• Transformers 4.51.2
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1