exceptions_exp2_swap_require_to_carry_40817

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

• Loss: 3.5533
• Accuracy: 0.3699

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: 40817
• 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
4.8165	0.2911	1000	0.2568	4.7391
4.3333	0.5822	2000	0.2995	4.2766
4.1412	0.8733	3000	0.3155	4.0926
3.9991	1.1642	4000	0.3250	3.9883
3.933	1.4553	5000	0.3325	3.9106
3.8824	1.7464	6000	0.3375	3.8533
3.7539	2.0373	7000	0.3419	3.8091
3.7421	2.3284	8000	0.3446	3.7809
3.7406	2.6195	9000	0.3474	3.7504
3.7255	2.9106	10000	0.3499	3.7260
3.6348	3.2014	11000	0.3514	3.7142
3.6457	3.4925	12000	0.3534	3.6941
3.6447	3.7837	13000	0.3549	3.6771
3.539	4.0745	14000	0.3566	3.6698
3.55	4.3656	15000	0.3576	3.6586
3.5796	4.6567	16000	0.3590	3.6455
3.576	4.9478	17000	0.3600	3.6317
3.4957	5.2387	18000	0.3602	3.6345
3.5063	5.5298	19000	0.3615	3.6241
3.5306	5.8209	20000	0.3622	3.6127
3.4346	6.1118	21000	0.3626	3.6142
3.472	6.4029	22000	0.3633	3.6078
3.4887	6.6940	23000	0.3639	3.5991
3.4872	6.9851	24000	0.3648	3.5891
3.4348	7.2760	25000	0.3648	3.5988
3.4425	7.5671	26000	0.3653	3.5901
3.4498	7.8582	27000	0.3661	3.5791
3.3936	8.1490	28000	0.3660	3.5864
3.3962	8.4401	29000	0.3664	3.5809
3.4249	8.7313	30000	0.3671	3.5757
3.3161	9.0221	31000	0.3672	3.5793
3.387	9.3132	32000	0.3674	3.5782
3.3924	9.6043	33000	0.3682	3.5694
3.4094	9.8954	34000	0.3685	3.5596
3.3295	10.1863	35000	0.3683	3.5720
3.362	10.4774	36000	0.3686	3.5671
3.3768	10.7685	37000	0.3692	3.5588
3.2931	11.0594	38000	0.3691	3.5665
3.3272	11.3505	39000	0.3697	3.5626
3.3696	11.6416	40000	0.3699	3.5533
3.3642	11.9327	41000	0.3705	3.5477
3.3012	12.2236	42000	0.3701	3.5630
3.3354	12.5147	43000	0.3706	3.5529
3.3443	12.8058	44000	0.3708	3.5489
3.2808	13.0966	45000	0.3704	3.5610
3.2958	13.3878	46000	0.3707	3.5527
3.3062	13.6789	47000	0.3714	3.5490
3.3443	13.9700	48000	0.3718	3.5406
3.272	14.2608	49000	0.3708	3.5555
3.2966	14.5519	50000	0.3717	3.5500
3.3184	14.8430	51000	0.3719	3.5418
3.2363	15.1339	52000	0.3713	3.5552
3.2768	15.4250	53000	0.3714	3.5532
3.2983	15.7161	54000	0.3721	3.5444
3.2505	16.0070	55000	0.3717	3.5532
3.2394	16.2981	56000	0.3719	3.5511
3.2745	16.5892	57000	0.3723	3.5431
3.2797	16.8803	58000	0.3730	3.5383
3.225	17.1712	59000	0.3721	3.5553
3.2534	17.4623	60000	0.3726	3.5480
3.2683	17.7534	61000	0.3733	3.5342
3.1819	18.0442	62000	0.3724	3.5529
3.2276	18.3354	63000	0.3727	3.5463
3.2585	18.6265	64000	0.3731	3.5394
3.2731	18.9176	65000	0.3736	3.5341
3.1994	19.2084	66000	0.3728	3.5523
3.2267	19.4995	67000	0.3731	3.5473
3.2469	19.7906	68000	0.3736	3.5351
3.1667	20.0815	69000	0.3732	3.5486
3.2161	20.3726	70000	0.3732	3.5458
3.2206	20.6637	71000	0.3739	3.5391
3.2496	20.9548	72000	0.3743	3.5302
3.1786	21.2457	73000	0.3733	3.5517
3.2064	21.5368	74000	0.3737	3.5392
3.224	21.8279	75000	0.3741	3.5349
3.1679	22.1188	76000	0.3733	3.5520
3.1935	22.4099	77000	0.3735	3.5461
3.2103	22.7010	78000	0.3736	3.5389
3.2216	22.9921	79000	0.3745	3.5328
3.1856	23.2830	80000	0.3737	3.5483
3.1841	23.5741	81000	3.5561	0.3735
3.2075	23.8652	82000	3.5405	0.3741
3.141	24.1563	83000	3.5551	0.3733
3.1806	24.4474	84000	3.5472	0.3741
3.2036	24.7385	85000	3.5373	0.3745
3.1092	25.0294	86000	3.5521	0.3740
3.1682	25.3205	87000	3.5488	0.3741
3.1867	25.6116	88000	3.5393	0.3748
3.2031	25.9027	89000	3.5329	0.3748
3.1405	26.1936	90000	3.5546	0.3740
3.1605	26.4847	91000	3.5474	0.3742
3.1786	26.7758	92000	3.5363	0.3749
3.1217	27.0667	93000	3.5516	0.3744
3.1474	27.3578	94000	3.5476	0.3743
3.1783	27.6489	95000	3.5408	0.3748
3.1755	27.9400	96000	3.5334	0.3752
3.1254	28.2308	97000	3.5512	0.3743
3.16	28.5219	98000	3.5444	0.3746
3.1722	28.8131	99000	3.5362	0.3754
3.0868	29.1039	100000	3.5495	0.3747

Framework versions

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