learning_rate: 1e-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 128
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
lr_scheduler_warmup_ratio: 0.1
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
18.1163	0.0055	50	5.3796
17.8993	0.0109	100	5.0878
16.9877	0.0164	150	4.9054
16.6951	0.0219	200	4.7595
17.1956	0.0273	250	nan
15.8726	0.0328	300	4.3051
15.8035	0.0382	350	4.1517
15.7724	0.0437	400	4.1179
15.2753	0.0492	450	4.0127
15.3521	0.0546	500	3.9190
14.8835	0.0601	550	3.9637
14.0689	0.0656	600	3.8433
14.8118	0.0710	650	3.6712
14.2018	0.0765	700	3.6710
13.9958	0.0819	750	3.6602
13.7916	0.0874	800	3.5779
13.5979	0.0929	850	3.5067
13.4006	0.0983	900	3.6269
13.5289	0.1038	950	3.4551
13.8164	0.1093	1000	3.4879
13.4817	0.1147	1050	3.3742
13.3695	0.1202	1100	3.4005
13.5545	0.1256	1150	nan
13.302	0.1311	1200	3.4059
12.5605	0.1366	1250	3.2717
13.0459	0.1420	1300	3.2960
12.2962	0.1475	1350	3.3724
12.8443	0.1530	1400	3.4170
12.8099	0.1584	1450	3.2809
12.7225	0.1639	1500	3.2522
12.3512	0.1694	1550	3.2800
12.1792	0.1748	1600	3.1975
12.1118	0.1803	1650	3.3046
12.6203	0.1857	1700	3.1530
12.1681	0.1912	1750	3.2485
12.1655	0.1967	1800	3.1499
12.2293	0.2021	1850	nan
13.0211	0.2076	1900	3.1991
12.4299	0.2131	1950	3.0704
12.6538	0.2185	2000	3.1030
12.6587	0.2240	2050	3.0889
12.2565	0.2294	2100	3.1421
12.5572	0.2349	2150	3.1872
12.2115	0.2404	2200	3.0676
12.0915	0.2458	2250	3.1293
12.219	0.2513	2300	3.0100
11.6999	0.2568	2350	3.0791
11.6684	0.2622	2400	2.9914
12.3173	0.2677	2450	3.0561
11.422	0.2731	2500	3.0956
11.5678	0.2786	2550	3.0362
12.0914	0.2841	2600	3.1488
11.7267	0.2895	2650	3.0109
11.8665	0.2950	2700	3.0516
11.8245	0.3005	2750	nan
11.5481	0.3059	2800	3.0401
11.9585	0.3114	2850	3.0800
11.539	0.3169	2900	3.0259
12.1617	0.3223	2950	2.9229
11.5909	0.3278	3000	nan
10.9824	0.3332	3050	2.7986
11.6864	0.3387	3100	2.9532
11.8222	0.3442	3150	2.9858
11.6286	0.3496	3200	2.9151
11.8334	0.3551	3250	2.9736
11.2803	0.3606	3300	2.8687
11.6026	0.3660	3350	2.8898
11.1204	0.3715	3400	2.8825
11.423	0.3769	3450	2.9872
11.4879	0.3824	3500	nan
11.573	0.3879	3550	2.8579
12.4166	0.3933	3600	2.9434
11.2353	0.3988	3650	2.8829
11.4	0.4043	3700	2.9354
10.9686	0.4097	3750	2.9237
11.4042	0.4152	3800	3.0190
11.8554	0.4207	3850	2.9271
11.9632	0.4261	3900	3.0376
11.1689	0.4316	3950	2.8588
11.4857	0.4370	4000	3.0363
11.4006	0.4425	4050	nan
10.7614	0.4480	4100	2.9261
11.5512	0.4534	4150	2.8923
11.5332	0.4589	4200	2.9156
11.0784	0.4644	4250	2.8987
10.7075	0.4698	4300	2.8653
11.6934	0.4753	4350	2.9077
11.4598	0.4807	4400	2.8124
11.1979	0.4862	4450	2.9331
10.9605	0.4917	4500	2.8848
11.1071	0.4971	4550	2.8974
11.2478	0.5026	4600	2.8152
11.0701	0.5081	4650	nan
11.6265	0.5135	4700	2.7508
11.4536	0.5190	4750	2.8445
11.135	0.5244	4800	nan
10.6803	0.5299	4850	2.8144
11.5174	0.5354	4900	nan
10.9389	0.5408	4950	2.8543
10.7885	0.5463	5000	2.8492
11.4897	0.5518	5050	2.8976
11.0512	0.5572	5100	2.8881
11.0947	0.5627	5150	2.8912
10.8325	0.5682	5200	2.8968
11.0708	0.5736	5250	2.8654
10.7451	0.5791	5300	2.8796
11.1151	0.5845	5350	2.8472
11.1104	0.5900	5400	2.7230
10.8407	0.5955	5450	2.8088
11.3176	0.6009	5500	2.8639
10.667	0.6064	5550	2.8375
11.2007	0.6119	5600	2.7334
10.8507	0.6173	5650	2.9190
10.587	0.6228	5700	2.7530
10.9799	0.6282	5750	2.8033
10.9086	0.6337	5800	2.8213
11.3728	0.6392	5850	2.9052
10.8764	0.6446	5900	2.7960
10.801	0.6501	5950	2.7806
10.9585	0.6556	6000	2.8511
10.3226	0.6610	6050	2.8546
11.29	0.6665	6100	2.7878
11.0151	0.6719	6150	2.7628
10.9404	0.6774	6200	2.7391
10.6352	0.6829	6250	2.8001
10.8923	0.6883	6300	2.7758
11.1026	0.6938	6350	2.7681
10.9816	0.6993	6400	2.8254
10.8846	0.7047	6450	2.7311
11.5027	0.7102	6500	2.8385
10.9822	0.7157	6550	2.9160
10.2748	0.7211	6600	2.7859
10.8561	0.7266	6650	2.7536
11.0482	0.7320	6700	2.7821
10.7278	0.7375	6750	2.7903
11.1712	0.7430	6800	2.8584
11.2571	0.7484	6850	2.8279
11.1056	0.7539	6900	2.7671
10.6717	0.7594	6950	nan
11.6541	0.7648	7000	2.7546
10.9426	0.7703	7050	nan
10.9949	0.7757	7100	2.8047
10.5327	0.7812	7150	2.7311
10.6881	0.7867	7200	2.6669
10.6447	0.7921	7250	2.7921
10.6472	0.7976	7300	2.7847
10.8787	0.8031	7350	2.8591
10.2926	0.8085	7400	nan
10.2332	0.8140	7450	2.7988
11.0911	0.8194	7500	2.7928
10.3863	0.8249	7550	2.7478
11.0595	0.8304	7600	2.7977
11.0389	0.8358	7650	2.8220
10.4349	0.8413	7700	2.7518
10.9154	0.8468	7750	2.8029
11.232	0.8522	7800	2.7387
10.7908	0.8577	7850	2.6633
10.629	0.8632	7900	2.7734
10.7351	0.8686	7950	2.7511
10.7233	0.8741	8000	2.7507
10.5188	0.8795	8050	2.8178
10.3795	0.8850	8100	2.7701
10.3011	0.8905	8150	2.7541
10.9437	0.8959	8200	2.7943
10.4525	0.9014	8250	2.7283
10.832	0.9069	8300	2.6824
10.4682	0.9123	8350	2.7700
10.8953	0.9178	8400	2.7677
10.7738	0.9232	8450	2.8578
10.9378	0.9287	8500	2.7341
11.4643	0.9342	8550	2.7165
10.8033	0.9396	8600	2.7774
10.4552	0.9451	8650	2.6508
10.6864	0.9506	8700	2.8617
11.5173	0.9560	8750	2.6617
10.8722	0.9615	8800	2.6334
10.4516	0.9669	8850	2.8097
10.9324	0.9724	8900	2.7791
10.3205	0.9779	8950	2.6963
10.6569	0.9833	9000	2.7603
10.8572	0.9888	9050	nan
11.0403	0.9943	9100	2.7983
11.1694	0.9997	9150	2.7260

Framework versions

Transformers 4.47.1
Pytorch 2.5.1+cu124
Datasets 3.1.0
Tokenizers 0.21.0

BounharAbdelaziz/xlm-roberta-large-bs-16-lr-1e-05-ep-1-wp-0.1-gacc-8-gnm-1.0-FP16-mx-512-v0.1

xlm-roberta-large-bs-16-lr-1e-05-ep-1-wp-0.1-gacc-8-gnm-1.0-FP16-mx-512-v0.1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

BounharAbdelaziz/
xlm-roberta-large-bs-16-lr-1e-05-ep-1-wp-0.1-gacc-8-gnm-1.0-FP16-mx-512-v0.1