exceptions_exp2_swap_0.7_last_to_hit_2128

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

• Loss: 3.5661
• Accuracy: 0.3684

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: 2128
• 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	Accuracy	Validation Loss
4.8576	0.2915	1000	0.2522	4.7695
4.3443	0.5830	2000	0.2983	4.2924
4.1518	0.8745	3000	0.3142	4.1072
4.0086	1.1659	4000	0.3240	3.9998
3.9388	1.4574	5000	0.3306	3.9221
3.8814	1.7489	6000	0.3354	3.8675
3.7671	2.0402	7000	0.3399	3.8242
3.7644	2.3317	8000	0.3431	3.7926
3.747	2.6233	9000	0.3455	3.7630
3.7255	2.9148	10000	0.3481	3.7396
3.651	3.2061	11000	0.3495	3.7249
3.6587	3.4976	12000	0.3518	3.7051
3.6547	3.7891	13000	0.3533	3.6885
3.5439	4.0805	14000	0.3549	3.6792
3.5683	4.3720	15000	0.3561	3.6706
3.5995	4.6635	16000	0.3568	3.6583
3.5863	4.9550	17000	0.3585	3.6428
3.5009	5.2463	18000	0.3588	3.6434
3.5279	5.5378	19000	0.3597	3.6353
3.54	5.8293	20000	0.3607	3.6223
3.4569	6.1207	21000	0.3612	3.6293
3.4702	6.4122	22000	0.3617	3.6178
3.5007	6.7037	23000	0.3625	3.6082
3.5039	6.9952	24000	0.3635	3.5988
3.4572	7.2866	25000	0.3632	3.6083
3.4556	7.5781	26000	0.3642	3.6002
3.4615	7.8696	27000	0.3647	3.5909
3.39	8.1609	28000	0.3646	3.5974
3.4242	8.4524	29000	0.3651	3.5929
3.432	8.7439	30000	0.3657	3.5840
3.3399	9.0353	31000	0.3659	3.5876
3.3765	9.3268	32000	0.3661	3.5871
3.4105	9.6183	33000	0.3663	3.5772
3.417	9.9098	34000	0.3671	3.5714
3.352	10.2011	35000	0.3668	3.5833
3.3865	10.4927	36000	0.3672	3.5761
3.3949	10.7842	37000	0.3681	3.5677
3.2997	11.0755	38000	0.3675	3.5746
3.3415	11.3670	39000	0.3679	3.5741
3.3834	11.6585	40000	0.3684	3.5661
3.3817	11.9500	41000	0.3690	3.5601
3.3111	12.2414	42000	0.3687	3.5719
3.3478	12.5329	43000	0.3688	3.5658
3.3542	12.8244	44000	0.3696	3.5546
3.2715	13.1157	45000	0.3690	3.5696
3.311	13.4072	46000	0.3694	3.5655
3.3541	13.6988	47000	0.3700	3.5542
3.3397	13.9903	48000	0.3704	3.5493
3.286	14.2816	49000	0.3698	3.5646
3.321	14.5731	50000	0.3701	3.5602
3.3357	14.8646	51000	0.3705	3.5500
3.2628	15.1560	52000	0.3703	3.5641
3.2999	15.4475	53000	0.3704	3.5586
3.302	15.7390	54000	0.3710	3.5525
3.2155	16.0303	55000	0.3703	3.5637
3.2662	16.3218	56000	0.3706	3.5609
3.2923	16.6133	57000	0.3711	3.5522
3.3144	16.9049	58000	0.3717	3.5454
3.2414	17.1962	59000	0.3708	3.5629
3.2755	17.4877	60000	0.3714	3.5533
3.2902	17.7792	61000	0.3717	3.5472
3.1978	18.0705	62000	0.3713	3.5624
3.2489	18.3621	63000	0.3716	3.5541
3.2619	18.6536	64000	0.3720	3.5469
3.2906	18.9451	65000	0.3722	3.5420
3.2252	19.2364	66000	0.3713	3.5582
3.2384	19.5279	67000	0.3721	3.5521
3.2649	19.8194	68000	0.3721	3.5461
3.1935	20.1108	69000	0.3715	3.5583
3.2157	20.4023	70000	0.3718	3.5591
3.2433	20.6938	71000	0.3724	3.5468
3.2726	20.9853	72000	0.3729	3.5407
3.2125	21.2766	73000	0.3719	3.5581
3.2289	21.5682	74000	0.3725	3.5494
3.2557	21.8597	75000	0.3727	3.5433
3.1787	22.1510	76000	0.3720	3.5586
3.2007	22.4425	77000	0.3724	3.5494
3.2251	22.7340	78000	0.3729	3.5456
3.149	23.0254	79000	0.3724	3.5563
3.1792	23.3169	80000	0.3723	3.5551
3.1916	23.6084	81000	3.5617	0.3723
3.1952	23.8999	82000	3.5500	0.3727
3.1583	24.1915	83000	3.5620	0.3723
3.1903	24.4830	84000	3.5512	0.3730
3.2194	24.7745	85000	3.5449	0.3731
3.1467	25.0659	86000	3.5610	0.3726
3.1807	25.3574	87000	3.5508	0.3728
3.1973	25.6489	88000	3.5461	0.3733
3.2084	25.9404	89000	3.5425	0.3735
3.1694	26.2318	90000	3.5613	0.3726
3.18	26.5233	91000	3.5519	0.3732
3.2064	26.8148	92000	3.5454	0.3734
3.126	27.1061	93000	3.5600	0.3730
3.1559	27.3976	94000	3.5542	0.3731
3.1835	27.6891	95000	3.5473	0.3735
3.2082	27.9806	96000	3.5390	0.3742
3.1359	28.2720	97000	3.5580	0.3731
3.1698	28.5635	98000	3.5524	0.3733
3.1724	28.8550	99000	3.5459	0.3736
3.1241	29.1463	100000	3.5589	0.3731
3.1388	29.4378	101000	3.5580	0.3732
3.1517	29.7294	102000	3.5468	0.3740
3.0817	30.0207	103000	3.5568	0.3734
3.1331	30.3122	104000	3.5567	0.3732
3.1452	30.6037	105000	3.5513	0.3737
3.1639	30.8952	106000	3.5459	0.3741
3.1105	31.1866	107000	3.5622	0.3731
3.1216	31.4781	108000	3.5537	0.3735
3.1607	31.7696	109000	3.5512	0.3739
3.0766	32.0609	110000	3.5614	0.3734
3.1082	32.3524	111000	3.5586	0.3740
3.1309	32.6439	112000	3.5503	0.3738
3.1492	32.9355	113000	3.5484	0.3744
3.1007	33.2268	114000	3.5572	0.3737
3.1038	33.5183	115000	3.5560	0.3738
3.1385	33.8098	116000	3.5470	0.3744

Framework versions

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