exceptions_exp2_swap_last_to_drop_1032

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

• Loss: 3.5702
• Accuracy: 0.3682

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
4.8404	0.2915	1000	4.7573	0.2540
4.3478	0.5830	2000	4.2933	0.2982
4.1542	0.8744	3000	4.1071	0.3143
3.9937	1.1659	4000	3.9985	0.3239
3.9404	1.4573	5000	3.9233	0.3306
3.8849	1.7488	6000	3.8651	0.3356
3.7673	2.0402	7000	3.8213	0.3403
3.7576	2.3317	8000	3.7921	0.3432
3.7297	2.6232	9000	3.7627	0.3460
3.744	2.9147	10000	3.7377	0.3480
3.6421	3.2061	11000	3.7239	0.3501
3.6474	3.4976	12000	3.7071	0.3518
3.6607	3.7890	13000	3.6881	0.3536
3.5493	4.0804	14000	3.6798	0.3547
3.5868	4.3719	15000	3.6686	0.3556
3.5868	4.6634	16000	3.6551	0.3571
3.5794	4.9549	17000	3.6443	0.3584
3.5133	5.2463	18000	3.6461	0.3586
3.5279	5.5378	19000	3.6332	0.3597
3.5482	5.8293	20000	3.6252	0.3606
3.4488	6.1207	21000	3.6296	0.3608
3.4847	6.4121	22000	3.6229	0.3617
3.4936	6.7036	23000	3.6110	0.3626
3.503	6.9951	24000	3.6029	0.3632
3.4459	7.2865	25000	3.6099	0.3633
3.4611	7.5780	26000	3.6014	0.3639
3.4639	7.8695	27000	3.5885	0.3646
3.3989	8.1609	28000	3.5995	0.3647
3.4142	8.4524	29000	3.5954	0.3650
3.4447	8.7438	30000	3.5830	0.3657
3.3361	9.0353	31000	3.5873	0.3658
3.3878	9.3267	32000	3.5883	0.3659
3.3991	9.6182	33000	3.5787	0.3664
3.4152	9.9097	34000	3.5716	0.3672
3.3479	10.2011	35000	3.5840	0.3668
3.3672	10.4926	36000	3.5788	0.3672
3.4002	10.7841	37000	3.5728	0.3679
3.3029	11.0755	38000	3.5797	0.3675
3.3538	11.3670	39000	3.5754	0.3679
3.3687	11.6584	40000	3.5702	0.3682
3.3743	11.9499	41000	3.5575	0.3690
3.3106	12.2413	42000	3.5730	0.3682
3.3557	12.5328	43000	3.5691	0.3687
3.3566	12.8243	44000	3.5601	0.3695
3.2706	13.1157	45000	3.5751	0.3684
3.3143	13.4072	46000	3.5689	0.3690
3.3453	13.6987	47000	3.5596	0.3694
3.3437	13.9901	48000	3.5515	0.3701
3.2881	14.2816	49000	3.5667	0.3695
3.3216	14.5730	50000	3.5590	0.3699
3.3344	14.8645	51000	3.5547	0.3704
3.2518	15.1559	52000	3.5646	0.3699
3.2999	15.4474	53000	3.5645	0.3697
3.2995	15.7389	54000	3.5512	0.3708
3.2182	16.0303	55000	3.5614	0.3705
3.268	16.3218	56000	3.5613	0.3705
3.2985	16.6133	57000	3.5542	0.3710
3.3078	16.9047	58000	3.5474	0.3713
3.2442	17.1962	59000	3.5656	0.3706
3.2631	17.4876	60000	3.5542	0.3712
3.2891	17.7791	61000	3.5467	0.3714
3.1964	18.0705	62000	3.5665	0.3711
3.2479	18.3620	63000	3.5632	0.3711
3.2704	18.6535	64000	3.5506	0.3716
3.2754	18.9450	65000	3.5441	0.3722
3.2223	19.2364	66000	3.5614	0.3715
3.2462	19.5279	67000	3.5524	0.3719
3.2614	19.8193	68000	3.5479	0.3721
3.172	20.1108	69000	3.5637	0.3715
3.2315	20.4022	70000	3.5592	0.3715
3.2457	20.6937	71000	3.5489	0.3722
3.2639	20.9852	72000	3.5472	0.3726
3.211	21.2766	73000	3.5613	0.3715
3.2372	21.5681	74000	3.5497	0.3722
3.253	21.8596	75000	3.5441	0.3725
3.1812	22.1510	76000	3.5588	0.3723
3.2157	22.4425	77000	3.5557	0.3721
3.2323	22.7339	78000	3.5459	0.3729
3.1398	23.0254	79000	3.5602	0.3721
3.1957	23.3168	80000	3.5591	0.3723
3.2137	23.6083	81000	3.5487	0.3726
3.2307	23.8998	82000	3.5483	0.3728
3.1643	24.1912	83000	3.5561	0.3727
3.2051	24.4827	84000	3.5567	0.3726
3.2181	24.7742	85000	3.5474	0.3731
3.1372	25.0656	86000	3.5627	0.3725
3.172	25.3571	87000	3.5571	0.3727
3.1886	25.6485	88000	3.5522	0.3731
3.2253	25.9400	89000	3.5430	0.3736
3.147	26.2314	90000	3.5625	0.3728
3.1795	26.5229	91000	3.5575	0.3730
3.2016	26.8144	92000	3.5438	0.3735
3.1211	27.1058	93000	3.5611	0.3728
3.1742	27.3973	94000	3.5582	0.3728
3.1787	27.6888	95000	3.5521	0.3730
3.2024	27.9802	96000	3.5435	0.3738
3.1213	28.2717	97000	3.5580	0.3733
3.1616	28.5631	98000	3.5547	0.3733
3.1816	28.8546	99000	3.5441	0.3736
3.1166	29.1460	100000	3.5581	0.3732
3.163	29.4375	101000	3.5542	0.3735
3.1574	29.7290	102000	3.5502	0.3739
3.0737	30.0204	103000	3.5608	0.3736
3.1441	30.3119	104000	3.5588	0.3737
3.1518	30.6034	105000	3.5535	0.3737
3.1703	30.8948	106000	3.5461	0.3742
3.1048	31.1863	107000	3.5579	0.3736
3.1429	31.4777	108000	3.5550	0.3738
3.152	31.7692	109000	3.5483	0.3742

Framework versions

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