learning_rate: 0.0006
train_batch_size: 16
eval_batch_size: 16
seed: 5039
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
4.8203	0.2915	1000	4.7392	0.2566
4.3473	0.5830	2000	4.2866	0.2991
4.1651	0.8745	3000	4.1095	0.3142
3.9992	1.1659	4000	4.0001	0.3235
3.9435	1.4574	5000	3.9233	0.3307
3.8911	1.7488	6000	3.8661	0.3360
3.7584	2.0402	7000	3.8226	0.3398
3.7642	2.3317	8000	3.7898	0.3433
3.7596	2.6232	9000	3.7602	0.3461
3.7319	2.9147	10000	3.7356	0.3483
3.6474	3.2061	11000	3.7239	0.3504
3.6495	3.4976	12000	3.7047	0.3523
3.6494	3.7891	13000	3.6867	0.3535
3.5329	4.0805	14000	3.6779	0.3550
3.5828	4.3719	15000	3.6684	0.3561
3.582	4.6634	16000	3.6553	0.3571
3.5847	4.9549	17000	3.6400	0.3584
3.5057	5.2463	18000	3.6449	0.3590
3.5258	5.5378	19000	3.6329	0.3598
3.5255	5.8293	20000	3.6216	0.3608
3.4553	6.1207	21000	3.6277	0.3611
3.4777	6.4122	22000	3.6168	0.3619
3.4888	6.7037	23000	3.6070	0.3627
3.4997	6.9952	24000	3.5979	0.3638
3.4323	7.2865	25000	3.6083	0.3634
3.4579	7.5780	26000	3.5968	0.3641
3.4603	7.8695	27000	3.5887	0.3651
3.3839	8.1609	28000	3.5978	0.3647
3.4169	8.4524	29000	3.5894	0.3653
3.4387	8.7439	30000	3.5817	0.3660
3.3232	9.0353	31000	3.5871	0.3659
3.3788	9.3268	32000	3.5860	0.3662
3.3911	9.6183	33000	3.5764	0.3670
3.4125	9.9098	34000	3.5701	0.3675
3.3375	10.2011	35000	3.5835	0.3668
3.3836	10.4926	36000	3.5743	0.3676
3.3932	10.7841	37000	3.5638	0.3681
3.3074	11.0755	38000	3.5768	0.3678
3.3583	11.3670	39000	3.5734	0.3683
3.3791	11.6585	40000	3.5632	0.3688
3.3749	11.9500	41000	3.5588	0.3689
3.3119	12.2414	42000	3.5705	0.3688
3.3277	12.5329	43000	3.5626	0.3692
3.3633	12.8243	44000	3.5570	0.3696
3.2781	13.1157	45000	3.5687	0.3692
3.3236	13.4072	46000	3.5630	0.3692
3.3421	13.6987	47000	3.5574	0.3700
3.3432	13.9902	48000	3.5446	0.3707
3.2882	14.2816	49000	3.5647	0.3697
3.3239	14.5731	50000	3.5541	0.3703
3.3207	14.8646	51000	3.5504	0.3707
3.2551	15.1559	52000	3.5627	0.3701
3.296	15.4474	53000	3.5577	0.3706
3.312	15.7389	54000	3.5495	0.3710
3.2027	16.0303	55000	3.5620	0.3707
3.255	16.3218	56000	3.5605	0.3706
3.2945	16.6133	57000	3.5496	0.3712
3.2974	16.9048	58000	3.5464	0.3716
3.2372	17.1962	59000	3.5634	0.3711
3.2599	17.4877	60000	3.5570	0.3711
3.2811	17.7792	61000	3.5470	0.3718
3.2008	18.0705	62000	3.5604	0.3713
3.2433	18.3620	63000	3.5601	0.3712
3.2701	18.6535	64000	3.5458	0.3718
3.2773	18.9450	65000	3.5419	0.3722
3.2228	19.2364	66000	3.5566	0.3714
3.2503	19.5279	67000	3.5490	0.3721
3.264	19.8194	68000	3.5443	0.3723
3.1876	20.1108	69000	3.5576	0.3718
3.216	20.4023	70000	3.5544	0.3720
3.2406	20.6938	71000	3.5447	0.3725
3.2618	20.9853	72000	3.5401	0.3729
3.2135	21.2766	73000	3.5549	0.3721
3.2337	21.5681	74000	3.5487	0.3724
3.2454	21.8596	75000	3.5397	0.3727
3.1749	22.1510	76000	3.5590	0.3721
3.2082	22.4425	77000	3.5505	0.3726
3.2295	22.7340	78000	3.5462	0.3726
3.1518	23.0254	79000	3.5585	0.3725
3.1787	23.3169	80000	3.5554	0.3726
3.2195	23.6083	81000	3.5461	0.3730
3.2287	23.8998	82000	3.5419	0.3733
3.1717	24.1912	83000	3.5572	0.3726
3.2054	24.4827	84000	3.5491	0.3729
3.209	24.7742	85000	3.5430	0.3733
3.133	25.0656	86000	3.5578	0.3728
3.1824	25.3571	87000	3.5551	0.3726
3.2034	25.6486	88000	3.5458	0.3735
3.2086	25.9401	89000	3.5394	0.3738
3.1537	26.2314	90000	3.5577	0.3729
3.1814	26.5229	91000	3.5524	0.3729
3.1851	26.8144	92000	3.5446	0.3738
3.1252	27.1058	93000	3.5634	0.3726
3.1465	27.3973	94000	3.5563	0.3731
3.1813	27.6888	95000	3.5471	0.3737
3.2031	27.9803	96000	3.5390	0.3742
3.1192	28.2717	97000	3.5576	0.3732
3.1647	28.5632	98000	3.5537	0.3734
3.1856	28.8547	99000	3.5434	0.3740
3.1171	29.1460	100000	3.5632	0.3729
3.1668	29.4375	101000	3.5541	0.3735
3.1795	29.7290	102000	3.5460	0.3738
3.0824	30.0204	103000	3.5576	0.3735
3.1345	30.3119	104000	3.5616	0.3734
3.1473	30.6034	105000	3.5517	0.3740
3.1743	30.8949	106000	3.5437	0.3744
3.1091	31.1863	107000	3.5620	0.3733
3.125	31.4778	108000	3.5538	0.3736
3.1521	31.7693	109000	3.5440	0.3741
3.0811	32.0606	110000	3.5604	0.3735
3.1071	32.3521	111000	3.5583	0.3737
3.1362	32.6436	112000	3.5527	0.3741
3.1492	32.9351	113000	3.5455	0.3744
3.091	33.2265	114000	3.5611	0.3737
3.118	33.5180	115000	3.5540	0.3739
3.1403	33.8095	116000	3.5469	0.3744

Framework versions

Transformers 4.55.2
Pytorch 2.8.0+cu128
Datasets 4.0.0
Tokenizers 0.21.4

craa/exceptions_exp2_swap_0.3_resemble_to_drop_5039

exceptions_exp2_swap_0.3_resemble_to_drop_5039

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

craa/
exceptions_exp2_swap_0.3_resemble_to_drop_5039