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
4.8301	0.2915	1000	4.7704	0.2528
4.3456	0.5830	2000	4.2872	0.2984
4.1521	0.8745	3000	4.1024	0.3140
3.9949	1.1659	4000	4.0009	0.3233
3.9441	1.4574	5000	3.9235	0.3302
3.8985	1.7488	6000	3.8644	0.3360
3.7581	2.0402	7000	3.8217	0.3404
3.7603	2.3317	8000	3.7908	0.3432
3.763	2.6232	9000	3.7613	0.3459
3.7299	2.9147	10000	3.7354	0.3483
3.6445	3.2061	11000	3.7236	0.3502
3.6587	3.4976	12000	3.7050	0.3519
3.6521	3.7891	13000	3.6892	0.3536
3.5561	4.0805	14000	3.6801	0.3545
3.5999	4.3719	15000	3.6690	0.3559
3.5846	4.6634	16000	3.6558	0.3566
3.5964	4.9549	17000	3.6426	0.3584
3.5139	5.2463	18000	3.6446	0.3588
3.5232	5.5378	19000	3.6331	0.3596
3.5372	5.8293	20000	3.6212	0.3604
3.4523	6.1207	21000	3.6265	0.3609
3.4854	6.4122	22000	3.6194	0.3616
3.4957	6.7037	23000	3.6089	0.3623
3.5034	6.9952	24000	3.5995	0.3631
3.429	7.2865	25000	3.6098	0.3633
3.4553	7.5780	26000	3.5984	0.3640
3.4653	7.8695	27000	3.5889	0.3646
3.3813	8.1609	28000	3.6000	0.3646
3.4265	8.4524	29000	3.5922	0.3649
3.4375	8.7439	30000	3.5845	0.3657
3.3381	9.0353	31000	3.5909	0.3655
3.3871	9.3268	32000	3.5885	0.3659
3.4087	9.6183	33000	3.5808	0.3665
3.4222	9.9098	34000	3.5707	0.3672
3.3548	10.2011	35000	3.5860	0.3664
3.383	10.4926	36000	3.5755	0.3672
3.3892	10.7841	37000	3.5666	0.3680
3.3137	11.0755	38000	3.5776	0.3676
3.3517	11.3670	39000	3.5727	0.3677
3.3591	11.6585	40000	3.5666	0.3683
3.3902	11.9500	41000	3.5585	0.3688
3.3036	12.2414	42000	3.5734	0.3683
3.3369	12.5329	43000	3.5666	0.3688
3.366	12.8243	44000	3.5576	0.3696
3.2779	13.1157	45000	3.5723	0.3687
3.3221	13.4072	46000	3.5645	0.3694
3.3442	13.6987	47000	3.5572	0.3694
3.3482	13.9902	48000	3.5534	0.3704
3.2899	14.2816	49000	3.5633	0.3699
3.3161	14.5731	50000	3.5581	0.3702
3.3284	14.8646	51000	3.5514	0.3707
3.2422	15.1559	52000	3.5681	0.3698
3.2778	15.4474	53000	3.5617	0.3704
3.3088	15.7389	54000	3.5557	0.3707
3.2182	16.0303	55000	3.5605	0.3707
3.2781	16.3218	56000	3.5600	0.3708
3.2853	16.6133	57000	3.5519	0.3710
3.3068	16.9048	58000	3.5462	0.3713
3.2309	17.1962	59000	3.5635	0.3704
3.255	17.4877	60000	3.5555	0.3710
3.2922	17.7792	61000	3.5482	0.3717
3.2143	18.0705	62000	3.5590	0.3711
3.2406	18.3620	63000	3.5561	0.3715
3.261	18.6535	64000	3.5519	0.3716
3.2808	18.9450	65000	3.5413	0.3722
3.2087	19.2364	66000	3.5602	0.3715
3.2543	19.5279	67000	3.5509	0.3718
3.2676	19.8194	68000	3.5440	0.3721
3.1908	20.1108	69000	3.5582	0.3718
3.2165	20.4023	70000	3.5575	0.3714
3.2678	20.6938	71000	3.5434	0.3723
3.2547	20.9853	72000	3.5414	0.3729
3.2093	21.2766	73000	3.5579	0.3717
3.2363	21.5681	74000	3.5489	0.3726
3.251	21.8596	75000	3.5407	0.3730
3.1749	22.1510	76000	3.5583	0.3722
3.2053	22.4425	77000	3.5507	0.3727
3.2462	22.7340	78000	3.5476	0.3726
3.1365	23.0254	79000	3.5560	0.3724
3.1845	23.3169	80000	3.5549	0.3725
3.2172	23.6083	81000	3.5512	0.3726
3.2164	23.8998	82000	3.5422	0.3732
3.1599	24.1912	83000	3.5603	0.3723
3.1822	24.4827	84000	3.5506	0.3733
3.215	24.7742	85000	3.5432	0.3734
3.1304	25.0656	86000	3.5577	0.3726
3.1653	25.3571	87000	3.5531	0.3728
3.1943	25.6486	88000	3.5507	0.3732
3.2182	25.9401	89000	3.5376	0.3736
3.1537	26.2314	90000	3.5574	0.3729
3.1822	26.5229	91000	3.5504	0.3731
3.1982	26.8144	92000	3.5467	0.3734
3.1306	27.1058	93000	3.5603	0.3725
3.1605	27.3973	94000	3.5567	0.3731
3.1784	27.6888	95000	3.5488	0.3735
3.1932	27.9803	96000	3.5408	0.3740
3.1489	28.2717	97000	3.5550	0.3733
3.1628	28.5632	98000	3.5498	0.3735
3.1892	28.8547	99000	3.5434	0.3740
3.1303	29.1460	100000	3.5574	0.3730
3.1466	29.4375	101000	3.5546	0.3734
3.1736	29.7290	102000	3.5474	0.3738
3.0903	30.0204	103000	3.5597	0.3733
3.13	30.3119	104000	3.5551	0.3735
3.1387	30.6034	105000	3.5505	0.3739
3.1696	30.8949	106000	3.5436	0.3741
3.1077	31.1863	107000	3.5614	0.3737
3.1382	31.4778	108000	3.5522	0.3739
3.1474	31.7693	109000	3.5488	0.3742

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_last_to_push_40817

exceptions_exp2_swap_0.3_last_to_push_40817

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

craa/
exceptions_exp2_swap_0.3_last_to_push_40817