exceptions_exp2_swap_last_to_hit_1032

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

• Loss: 3.5658
• Accuracy: 0.3685

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: 1032
• 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
4.83	0.2915	1000	4.7503	0.2556
4.3487	0.5830	2000	4.2899	0.2987
4.1543	0.8744	3000	4.1056	0.3147
3.9911	1.1659	4000	3.9968	0.3237
3.9398	1.4573	5000	3.9224	0.3307
3.8836	1.7488	6000	3.8657	0.3356
3.7668	2.0402	7000	3.8221	0.3400
3.7561	2.3317	8000	3.7928	0.3432
3.7292	2.6232	9000	3.7625	0.3458
3.7437	2.9147	10000	3.7367	0.3482
3.6402	3.2061	11000	3.7230	0.3502
3.6466	3.4976	12000	3.7035	0.3519
3.6599	3.7890	13000	3.6856	0.3537
3.5479	4.0804	14000	3.6806	0.3547
3.5858	4.3719	15000	3.6666	0.3558
3.5845	4.6634	16000	3.6536	0.3572
3.578	4.9549	17000	3.6407	0.3585
3.5125	5.2463	18000	3.6449	0.3587
3.5265	5.5378	19000	3.6320	0.3597
3.5466	5.8293	20000	3.6207	0.3607
3.4463	6.1207	21000	3.6256	0.3613
3.4832	6.4121	22000	3.6191	0.3617
3.4922	6.7036	23000	3.6074	0.3626
3.5013	6.9951	24000	3.5991	0.3633
3.4426	7.2865	25000	3.6085	0.3633
3.4602	7.5780	26000	3.5973	0.3640
3.4615	7.8695	27000	3.5865	0.3646
3.3956	8.1609	28000	3.5989	0.3647
3.413	8.4524	29000	3.5927	0.3648
3.4431	8.7438	30000	3.5815	0.3658
3.3348	9.0353	31000	3.5883	0.3655
3.3866	9.3267	32000	3.5847	0.3661
3.3968	9.6182	33000	3.5755	0.3667
3.4135	9.9097	34000	3.5702	0.3669
3.3465	10.2011	35000	3.5808	0.3668
3.3631	10.4926	36000	3.5766	0.3672
3.3981	10.7841	37000	3.5665	0.3681
3.3	11.0755	38000	3.5771	0.3679
3.3515	11.3670	39000	3.5736	0.3680
3.3672	11.6584	40000	3.5658	0.3685
3.3718	11.9499	41000	3.5540	0.3695
3.3073	12.2413	42000	3.5704	0.3684
3.3535	12.5328	43000	3.5635	0.3692
3.3541	12.8243	44000	3.5564	0.3693
3.2681	13.1157	45000	3.5685	0.3689
3.3097	13.4072	46000	3.5683	0.3688
3.3431	13.6987	47000	3.5554	0.3697
3.341	13.9901	48000	3.5493	0.3702
3.2848	14.2816	49000	3.5660	0.3694
3.3197	14.5730	50000	3.5554	0.3699
3.3313	14.8645	51000	3.5512	0.3707
3.2493	15.1559	52000	3.5637	0.3701
3.2978	15.4474	53000	3.5571	0.3702
3.2974	15.7389	54000	3.5484	0.3709
3.2158	16.0303	55000	3.5590	0.3704
3.265	16.3218	56000	3.5562	0.3706
3.296	16.6133	57000	3.5524	0.3710
3.3052	16.9047	58000	3.5458	0.3714
3.2412	17.1962	59000	3.5643	0.3706
3.2611	17.4876	60000	3.5540	0.3713
3.2879	17.7791	61000	3.5441	0.3717
3.1949	18.0705	62000	3.5642	0.3711
3.2444	18.3620	63000	3.5598	0.3711
3.2682	18.6535	64000	3.5489	0.3718
3.2727	18.9450	65000	3.5421	0.3722
3.2203	19.2364	66000	3.5575	0.3715
3.2438	19.5279	67000	3.5506	0.3718
3.2597	19.8193	68000	3.5429	0.3723
3.17	20.1108	69000	3.5620	0.3714
3.228	20.4022	70000	3.5557	0.3718
3.2443	20.6937	71000	3.5472	0.3722
3.2616	20.9852	72000	3.5413	0.3730
3.2101	21.2766	73000	3.5560	0.3718
3.2343	21.5681	74000	3.5482	0.3722
3.2504	21.8596	75000	3.5410	0.3727
3.1783	22.1510	76000	3.5568	0.3720
3.214	22.4425	77000	3.5545	0.3722
3.2304	22.7339	78000	3.5450	0.3727
3.1368	23.0254	79000	3.5537	0.3723
3.1932	23.3168	80000	3.5587	0.3723
3.2115	23.6083	81000	3.5467	0.3730
3.2283	23.8998	82000	3.5388	0.3732
3.1603	24.1912	83000	3.5566	0.3726
3.204	24.4827	84000	3.5493	0.3729
3.2166	24.7742	85000	3.5451	0.3732
3.1337	25.0656	86000	3.5615	0.3723
3.1697	25.3571	87000	3.5582	0.3726
3.186	25.6485	88000	3.5479	0.3732
3.225	25.9400	89000	3.5423	0.3733
3.1453	26.2314	90000	3.5594	0.3724
3.1765	26.5229	91000	3.5534	0.3728
3.1994	26.8144	92000	3.5437	0.3736
3.1196	27.1058	93000	3.5598	0.3727
3.1731	27.3973	94000	3.5568	0.3730
3.1778	27.6888	95000	3.5493	0.3731
3.2006	27.9802	96000	3.5356	0.3739
3.1193	28.2717	97000	3.5558	0.3733
3.1601	28.5631	98000	3.5504	0.3735
3.181	28.8546	99000	3.5435	0.3737
3.1149	29.1460	100000	3.5611	0.3729
3.1605	29.4375	101000	3.5535	0.3734
3.1568	29.7290	102000	3.5481	0.3738
3.0728	30.0204	103000	3.5601	0.3732
3.1414	30.3119	104000	3.5585	0.3732
3.1489	30.6034	105000	3.5511	0.3737
3.1693	30.8948	106000	3.5428	0.3741
3.1037	31.1863	107000	3.5594	0.3735
3.1414	31.4777	108000	3.5558	0.3735
3.1507	31.7692	109000	3.5475	0.3741
3.0806	32.0606	110000	3.5622	0.3733
3.1082	32.3521	111000	3.5579	0.3736
3.13	32.6436	112000	3.5505	0.3740
3.1337	32.9351	113000	3.5461	0.3742
3.1073	33.2265	114000	3.5615	0.3731
3.1209	33.5180	115000	3.5552	0.3740
3.1315	33.8094	116000	3.5471	0.3744

Framework versions

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