exceptions_exp2_swap_0.3_last_to_carry_1032

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

• Loss: 3.5866
• Accuracy: 0.3653

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
4.8263	0.2915	1000	4.7561	0.2543
4.3556	0.5830	2000	4.2907	0.2979
4.1533	0.8745	3000	4.1087	0.3137
4.0097	1.1659	4000	4.0010	0.3234
3.9521	1.4574	5000	3.9255	0.3302
3.8928	1.7488	6000	3.8709	0.3358
3.7556	2.0402	7000	3.8230	0.3400
3.7544	2.3317	8000	3.7933	0.3429
3.7479	2.6232	9000	3.7640	0.3454
3.735	2.9147	10000	3.7387	0.3482
3.6496	3.2061	11000	3.7282	0.3502
3.6511	3.4976	12000	3.7073	0.3515
3.6628	3.7891	13000	3.6888	0.3533
3.5616	4.0805	14000	3.6832	0.3544
3.5725	4.3719	15000	3.6700	0.3557
3.5912	4.6634	16000	3.6564	0.3568
3.5847	4.9549	17000	3.6445	0.3581
3.5181	5.2463	18000	3.6455	0.3588
3.517	5.5378	19000	3.6369	0.3597
3.5497	5.8293	20000	3.6263	0.3605
3.4518	6.1207	21000	3.6255	0.3610
3.4895	6.4122	22000	3.6214	0.3613
3.4915	6.7037	23000	3.6125	0.3626
3.5125	6.9952	24000	3.6024	0.3631
3.433	7.2865	25000	3.6097	0.3630
3.4677	7.5780	26000	3.6010	0.3640
3.4745	7.8695	27000	3.5914	0.3645
3.3773	8.1609	28000	3.6035	0.3645
3.4187	8.4524	29000	3.5952	0.3649
3.4491	8.7439	30000	3.5866	0.3653
3.3357	9.0353	31000	3.5907	0.3658
3.3944	9.3268	32000	3.5890	0.3658
3.3953	9.6183	33000	3.5794	0.3664
3.4163	9.9098	34000	3.5711	0.3671
3.3529	10.2011	35000	3.5841	0.3666
3.3694	10.4926	36000	3.5781	0.3673
3.3886	10.7841	37000	3.5683	0.3678
3.3082	11.0755	38000	3.5818	0.3676
3.3497	11.3670	39000	3.5747	0.3677
3.3708	11.6585	40000	3.5697	0.3684
3.3746	11.9500	41000	3.5607	0.3688
3.3285	12.2414	42000	3.5739	0.3684
3.3466	12.5329	43000	3.5696	0.3689
3.3627	12.8243	44000	3.5617	0.3693
3.2851	13.1157	45000	3.5724	0.3685
3.314	13.4072	46000	3.5643	0.3691
3.3345	13.6987	47000	3.5575	0.3695
3.3462	13.9902	48000	3.5529	0.3701
3.2879	14.2816	49000	3.5685	0.3694
3.3282	14.5731	50000	3.5594	0.3699
3.3389	14.8646	51000	3.5524	0.3701
3.2624	15.1559	52000	3.5681	0.3699
3.2954	15.4474	53000	3.5608	0.3701
3.3077	15.7389	54000	3.5542	0.3706
3.2257	16.0303	55000	3.5629	0.3706
3.268	16.3218	56000	3.5614	0.3705
3.2886	16.6133	57000	3.5542	0.3710
3.3106	16.9048	58000	3.5465	0.3712
3.2317	17.1962	59000	3.5646	0.3705
3.2643	17.4877	60000	3.5554	0.3707
3.2942	17.7792	61000	3.5493	0.3716
3.2074	18.0705	62000	3.5614	0.3711
3.2424	18.3620	63000	3.5583	0.3712
3.2657	18.6535	64000	3.5530	0.3714
3.287	18.9450	65000	3.5433	0.3719
3.2333	19.2364	66000	3.5610	0.3710
3.2578	19.5279	67000	3.5527	0.3714
3.2633	19.8194	68000	3.5466	0.3720
3.1828	20.1108	69000	3.5630	0.3714
3.2285	20.4023	70000	3.5581	0.3714
3.2468	20.6938	71000	3.5478	0.3721
3.2602	20.9853	72000	3.5400	0.3726
3.2109	21.2766	73000	3.5569	0.3719
3.2269	21.5681	74000	3.5527	0.3721
3.2491	21.8596	75000	3.5441	0.3730
3.1702	22.1510	76000	3.5604	0.3717
3.2135	22.4425	77000	3.5526	0.3722
3.2283	22.7340	78000	3.5468	0.3725
3.1382	23.0254	79000	3.5586	0.3723
3.2013	23.3169	80000	3.5551	0.3722
3.2213	23.6083	81000	3.5502	0.3726
3.2379	23.8998	82000	3.5432	0.3731
3.1632	24.1912	83000	3.5579	0.3725
3.1967	24.4827	84000	3.5533	0.3725
3.2194	24.7742	85000	3.5446	0.3731
3.1466	25.0656	86000	3.5630	0.3723
3.1762	25.3571	87000	3.5580	0.3725
3.1993	25.6486	88000	3.5501	0.3728
3.2104	25.9401	89000	3.5378	0.3736
3.1599	26.2314	90000	3.5594	0.3727
3.1783	26.5229	91000	3.5543	0.3728
3.2142	26.8144	92000	3.5453	0.3732
3.125	27.1058	93000	3.5585	0.3726
3.1643	27.3973	94000	3.5565	0.3728
3.1817	27.6888	95000	3.5494	0.3734
3.1939	27.9803	96000	3.5401	0.3737
3.1343	28.2717	97000	3.5592	0.3727
3.1796	28.5632	98000	3.5520	0.3733
3.1855	28.8547	99000	3.5405	0.3740
3.1222	29.1460	100000	3.5605	0.3732
3.1475	29.4375	101000	3.5569	0.3733
3.1768	29.7290	102000	3.5468	0.3735
3.0977	30.0204	103000	3.5610	0.3730
3.1359	30.3119	104000	3.5589	0.3733
3.1579	30.6034	105000	3.5494	0.3737
3.1741	30.8949	106000	3.5429	0.3740
3.1262	31.1863	107000	3.5605	0.3731
3.1388	31.4778	108000	3.5541	0.3739
3.1571	31.7693	109000	3.5490	0.3739

Framework versions

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