exceptions_exp2_swap_0.7_last_to_carry_3591

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

• Loss: 3.5657
• 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: 3591
• 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
4.8169	0.2915	1000	4.7479	0.2558
4.3424	0.5830	2000	4.2855	0.2987
4.1686	0.8745	3000	4.1059	0.3142
3.9864	1.1659	4000	3.9998	0.3242
3.9402	1.4574	5000	3.9191	0.3310
3.8948	1.7489	6000	3.8637	0.3356
3.748	2.0402	7000	3.8184	0.3407
3.7534	2.3317	8000	3.7896	0.3435
3.7364	2.6233	9000	3.7613	0.3460
3.7264	2.9148	10000	3.7319	0.3487
3.6422	3.2061	11000	3.7230	0.3506
3.6442	3.4976	12000	3.7026	0.3522
3.6364	3.7891	13000	3.6843	0.3539
3.5504	4.0805	14000	3.6779	0.3550
3.5723	4.3720	15000	3.6667	0.3564
3.5804	4.6635	16000	3.6522	0.3574
3.5832	4.9550	17000	3.6395	0.3586
3.512	5.2463	18000	3.6418	0.3591
3.5318	5.5378	19000	3.6316	0.3601
3.5356	5.8293	20000	3.6186	0.3609
3.4467	6.1207	21000	3.6247	0.3614
3.4882	6.4122	22000	3.6152	0.3619
3.4965	6.7037	23000	3.6071	0.3628
3.4962	6.9952	24000	3.5981	0.3634
3.4305	7.2866	25000	3.6049	0.3634
3.4511	7.5781	26000	3.5968	0.3640
3.4666	7.8696	27000	3.5868	0.3652
3.3874	8.1609	28000	3.5998	0.3647
3.4115	8.4524	29000	3.5903	0.3653
3.4299	8.7439	30000	3.5818	0.3658
3.3315	9.0353	31000	3.5858	0.3658
3.3775	9.3268	32000	3.5866	0.3658
3.4005	9.6183	33000	3.5781	0.3665
3.4255	9.9098	34000	3.5674	0.3674
3.3457	10.2011	35000	3.5798	0.3670
3.3726	10.4927	36000	3.5749	0.3674
3.3847	10.7842	37000	3.5661	0.3681
3.3131	11.0755	38000	3.5783	0.3678
3.3377	11.3670	39000	3.5696	0.3680
3.3594	11.6585	40000	3.5657	0.3684
3.3811	11.9500	41000	3.5582	0.3688
3.3023	12.2414	42000	3.5707	0.3686
3.3467	12.5329	43000	3.5626	0.3690
3.3625	12.8244	44000	3.5569	0.3694
3.2802	13.1157	45000	3.5698	0.3690
3.3195	13.4072	46000	3.5683	0.3693
3.3255	13.6988	47000	3.5566	0.3698
3.354	13.9903	48000	3.5456	0.3705
3.289	14.2816	49000	3.5639	0.3696
3.3147	14.5731	50000	3.5589	0.3701
3.3246	14.8646	51000	3.5494	0.3704
3.2493	15.1560	52000	3.5653	0.3699
3.2837	15.4475	53000	3.5575	0.3703
3.2974	15.7390	54000	3.5486	0.3708
3.2157	16.0303	55000	3.5641	0.3703
3.2678	16.3218	56000	3.5584	0.3707
3.2864	16.6133	57000	3.5514	0.3709
3.2855	16.9049	58000	3.5439	0.3715
3.2363	17.1962	59000	3.5635	0.3708
3.2619	17.4877	60000	3.5563	0.3711
3.2721	17.7792	61000	3.5452	0.3717
3.2021	18.0705	62000	3.5592	0.3713
3.243	18.3621	63000	3.5571	0.3712
3.263	18.6536	64000	3.5481	0.3720
3.2827	18.9451	65000	3.5411	0.3722
3.2131	19.2364	66000	3.5573	0.3712
3.2353	19.5279	67000	3.5525	0.3716
3.2705	19.8194	68000	3.5444	0.3721
3.1942	20.1108	69000	3.5565	0.3716
3.2129	20.4023	70000	3.5520	0.3719
3.2504	20.6938	71000	3.5476	0.3722
3.2507	20.9853	72000	3.5436	0.3726
3.1994	21.2766	73000	3.5578	0.3721
3.23	21.5682	74000	3.5491	0.3725
3.2457	21.8597	75000	3.5442	0.3725
3.1795	22.1510	76000	3.5588	0.3722
3.2034	22.4425	77000	3.5502	0.3725
3.2219	22.7340	78000	3.5438	0.3729
3.1523	23.0254	79000	3.5575	0.3722
3.1811	23.3169	80000	3.5547	0.3723
3.208	23.6084	81000	3.5478	0.3727
3.2241	23.8999	82000	3.5404	0.3733
3.165	24.1912	83000	3.5621	0.3724
3.1885	24.4827	84000	3.5546	0.3725
3.2059	24.7743	85000	3.5419	0.3731
3.1424	25.0656	86000	3.5568	0.3727
3.1671	25.3571	87000	3.5555	0.3727
3.1964	25.6486	88000	3.5474	0.3733
3.2115	25.9401	89000	3.5401	0.3738
3.1475	26.2315	90000	3.5600	0.3725
3.1906	26.5230	91000	3.5488	0.3735
3.18	26.8145	92000	3.5440	0.3735
3.13	27.1058	93000	3.5603	0.3729
3.1542	27.3973	94000	3.5553	0.3730
3.1741	27.6888	95000	3.5482	0.3738
3.1896	27.9804	96000	3.5376	0.3743
3.1229	28.2717	97000	3.5594	0.3733
3.1614	28.5632	98000	3.5503	0.3734
3.1806	28.8547	99000	3.5421	0.3741
3.1143	29.1460	100000	3.5583	0.3732
3.1506	29.4376	101000	3.5549	0.3733
3.161	29.7291	102000	3.5479	0.3740
3.0934	30.0204	103000	3.5602	0.3732
3.1233	30.3119	104000	3.5615	0.3729
3.1449	30.6034	105000	3.5489	0.3738
3.1635	30.8949	106000	3.5419	0.3742
3.1048	31.1863	107000	3.5603	0.3734
3.129	31.4778	108000	3.5522	0.3737
3.1504	31.7693	109000	3.5462	0.3740
3.0826	32.0606	110000	3.5579	0.3737
3.1069	32.3521	111000	3.5607	0.3734
3.1411	32.6437	112000	3.5495	0.3739
3.1511	32.9352	113000	3.5463	0.3745
3.1037	33.2265	114000	3.5639	0.3735
3.1226	33.5180	115000	3.5573	0.3739
3.1375	33.8095	116000	3.5452	0.3744

Framework versions

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