exceptions_exp2_swap_require_to_drop_40817

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

• Loss: 3.5574
• Accuracy: 0.3696

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: 0.0006
• train_batch_size: 16
• eval_batch_size: 16
• seed: 40817
• gradient_accumulation_steps: 5
• total_train_batch_size: 80
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 50.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
4.8466	0.2911	1000	4.7718	0.2530
4.3445	0.5822	2000	4.2865	0.2989
4.1487	0.8733	3000	4.0988	0.3150
4.0036	1.1642	4000	3.9907	0.3252
3.9375	1.4553	5000	3.9139	0.3321
3.8856	1.7464	6000	3.8561	0.3370
3.7575	2.0373	7000	3.8129	0.3415
3.7454	2.3284	8000	3.7831	0.3444
3.7444	2.6195	9000	3.7538	0.3472
3.7281	2.9106	10000	3.7293	0.3496
3.6378	3.2014	11000	3.7164	0.3514
3.6481	3.4925	12000	3.6967	0.3532
3.6473	3.7837	13000	3.6785	0.3549
3.5417	4.0745	14000	3.6731	0.3561
3.5539	4.3656	15000	3.6625	0.3571
3.5818	4.6567	16000	3.6471	0.3586
3.5794	4.9478	17000	3.6342	0.3596
3.4993	5.2387	18000	3.6365	0.3597
3.5089	5.5298	19000	3.6259	0.3608
3.5332	5.8209	20000	3.6142	0.3620
3.4391	6.1118	21000	3.6161	0.3625
3.4754	6.4029	22000	3.6096	0.3632
3.4919	6.6940	23000	3.6009	0.3639
3.4905	6.9851	24000	3.5917	0.3647
3.4381	7.2760	25000	3.5992	0.3647
3.4452	7.5671	26000	3.5925	0.3653
3.4527	7.8582	27000	3.5826	0.3658
3.3971	8.1490	28000	3.5878	0.3659
3.3992	8.4401	29000	3.5827	0.3663
3.4274	8.7313	30000	3.5765	0.3669
3.3172	9.0221	31000	3.5817	0.3670
3.3913	9.3132	32000	3.5789	0.3674
3.3959	9.6043	33000	3.5726	0.3679
3.4142	9.8954	34000	3.5616	0.3683
3.3329	10.1863	35000	3.5742	0.3682
3.3651	10.4774	36000	3.5704	0.3683
3.3795	10.7685	37000	3.5627	0.3691
3.298	11.0594	38000	3.5693	0.3690
3.3304	11.3505	39000	3.5656	0.3694
3.3744	11.6416	40000	3.5574	0.3696
3.3672	11.9327	41000	3.5496	0.3704
3.3059	12.2236	42000	3.5653	0.3698
3.3392	12.5147	43000	3.5553	0.3704
3.3481	12.8058	44000	3.5504	0.3707
3.2857	13.0966	45000	3.5624	0.3702
3.2993	13.3878	46000	3.5573	0.3706
3.3103	13.6789	47000	3.5536	0.3709
3.3482	13.9700	48000	3.5429	0.3714
3.2756	14.2608	49000	3.5565	0.3706
3.3011	14.5519	50000	3.5488	0.3714
3.3204	14.8430	51000	3.5413	0.3720
3.2399	15.1339	52000	3.5585	0.3711
3.2816	15.4250	53000	3.5553	0.3712
3.302	15.7161	54000	3.5453	0.3720
3.2544	16.0070	55000	3.5562	0.3715
3.2429	16.2981	56000	3.5524	0.3719
3.2782	16.5892	57000	3.5473	0.3720
3.2818	16.8803	58000	3.5381	0.3729
3.2286	17.1712	59000	3.5554	0.3720
3.2575	17.4623	60000	3.5501	0.3723
3.2729	17.7534	61000	3.5366	0.3732
3.185	18.0442	62000	3.5547	0.3722
3.2309	18.3354	63000	3.5518	0.3726
3.2619	18.6265	64000	3.5419	0.3730
3.2775	18.9176	65000	3.5380	0.3732
3.2032	19.2084	66000	3.5546	0.3724
3.2309	19.4995	67000	3.5462	0.3729
3.2505	19.7906	68000	3.5360	0.3735
3.1711	20.0815	69000	3.5537	0.3727
3.2195	20.3726	70000	3.5469	0.3729
3.223	20.6637	71000	3.5413	0.3735
3.2548	20.9548	72000	3.5315	0.3741
3.1821	21.2457	73000	3.5512	0.3730
3.2097	21.5368	74000	3.5420	0.3737
3.228	21.8279	75000	3.5325	0.3740
3.1735	22.1188	76000	3.5555	0.3731
3.1979	22.4099	77000	3.5454	0.3735
3.2138	22.7010	78000	3.5391	0.3736
3.2247	22.9921	79000	3.5318	0.3743
3.1887	23.2830	80000	3.5504	0.3737
3.2066	23.5741	81000	3.5404	0.3741
3.2304	23.8652	82000	3.5321	0.3742
3.1437	24.1560	83000	3.5520	0.3736
3.182	24.4471	84000	3.5453	0.3741
3.2048	24.7382	85000	3.5380	0.3743
3.1086	25.0291	86000	3.5513	0.3736
3.1688	25.3202	87000	3.5474	0.3741
3.1861	25.6113	88000	3.5413	0.3747
3.2058	25.9024	89000	3.5308	0.3748
3.1451	26.1933	90000	3.5570	0.3735
3.1654	26.4844	91000	3.5434	0.3743
3.1848	26.7755	92000	3.5385	0.3747
3.1256	27.0664	93000	3.5511	0.3741
3.1486	27.3575	94000	3.5487	0.3742
3.1817	27.6486	95000	3.5418	0.3748
3.1771	27.9397	96000	3.5343	0.3749
3.1297	28.2306	97000	3.5509	0.3741
3.1628	28.5217	98000	3.5443	0.3748
3.1719	28.8128	99000	3.5371	0.3751
3.0902	29.1036	100000	3.5485	0.3745
3.1338	29.3947	101000	3.5492	0.3743
3.1602	29.6858	102000	3.5455	0.3749
3.1786	29.9769	103000	3.5311	0.3756
3.1097	30.2678	104000	3.5509	0.3746
3.1368	30.5589	105000	3.5421	0.3751
3.1502	30.8500	106000	3.5364	0.3753
3.0948	31.1409	107000	3.5537	0.3745
3.1172	31.4320	108000	3.5426	0.3750
3.145	31.7231	109000	3.5424	0.3752

Framework versions

• Transformers 4.55.2
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.21.4