exceptions_exp2_swap_require_to_push_1032

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

• Loss: 3.5566
• Accuracy: 0.3697

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
4.8368	0.2911	1000	4.7602	0.2537
4.3514	0.5822	2000	4.2851	0.2993
4.1537	0.8733	3000	4.0964	0.3155
3.984	1.1642	4000	3.9906	0.3249
3.9259	1.4553	5000	3.9168	0.3315
3.8785	1.7464	6000	3.8579	0.3372
3.749	2.0373	7000	3.8152	0.3414
3.7559	2.3284	8000	3.7820	0.3445
3.7391	2.6195	9000	3.7550	0.3473
3.7358	2.9106	10000	3.7286	0.3496
3.6391	3.2014	11000	3.7160	0.3513
3.6562	3.4925	12000	3.6969	0.3531
3.6513	3.7837	13000	3.6790	0.3549
3.5372	4.0745	14000	3.6702	0.3562
3.5803	4.3656	15000	3.6639	0.3569
3.5727	4.6567	16000	3.6463	0.3584
3.5822	4.9478	17000	3.6362	0.3596
3.4961	5.2387	18000	3.6362	0.3602
3.5195	5.5298	19000	3.6283	0.3612
3.5309	5.8209	20000	3.6137	0.3623
3.4547	6.1118	21000	3.6172	0.3626
3.4809	6.4029	22000	3.6108	0.3632
3.4881	6.6940	23000	3.6033	0.3639
3.5017	6.9851	24000	3.5924	0.3645
3.4179	7.2760	25000	3.5990	0.3646
3.4484	7.5671	26000	3.5905	0.3652
3.4755	7.8582	27000	3.5820	0.3662
3.3854	8.1490	28000	3.5893	0.3660
3.4074	8.4401	29000	3.5852	0.3665
3.4407	8.7313	30000	3.5738	0.3670
3.3176	9.0221	31000	3.5825	0.3673
3.3657	9.3132	32000	3.5808	0.3674
3.3947	9.6043	33000	3.5694	0.3681
3.423	9.8954	34000	3.5631	0.3686
3.3361	10.1863	35000	3.5769	0.3684
3.3663	10.4774	36000	3.5682	0.3686
3.3733	10.7685	37000	3.5601	0.3693
3.2945	11.0594	38000	3.5687	0.3689
3.3349	11.3505	39000	3.5672	0.3692
3.3615	11.6416	40000	3.5566	0.3697
3.3565	11.9327	41000	3.5500	0.3703
3.3014	12.2236	42000	3.5638	0.3699
3.329	12.5147	43000	3.5575	0.3702
3.3453	12.8058	44000	3.5471	0.3709
3.2699	13.0966	45000	3.5661	0.3700
3.3058	13.3878	46000	3.5592	0.3706
3.319	13.6789	47000	3.5469	0.3711
3.3464	13.9700	48000	3.5417	0.3716
3.2773	14.2608	49000	3.5590	0.3708
3.3011	14.5519	50000	3.5518	0.3713
3.3317	14.8430	51000	3.5420	0.3720
3.2458	15.1339	52000	3.5598	0.3714
3.2734	15.4250	53000	3.5522	0.3717
3.2856	15.7161	54000	3.5418	0.3725
3.2596	16.0070	55000	3.5502	0.3722
3.2503	16.2981	56000	3.5522	0.3719
3.2839	16.5892	57000	3.5469	0.3722
3.2964	16.8803	58000	3.5369	0.3729
3.2298	17.1712	59000	3.5528	0.3724
3.2623	17.4623	60000	3.5480	0.3724
3.2786	17.7534	61000	3.5377	0.3729
3.1831	18.0442	62000	3.5501	0.3725
3.2437	18.3354	63000	3.5502	0.3728
3.267	18.6265	64000	3.5376	0.3733
3.2656	18.9176	65000	3.5346	0.3734
3.2097	19.2084	66000	3.5505	0.3726
3.2486	19.4995	67000	3.5444	0.3729
3.2669	19.7906	68000	3.5358	0.3736
3.1694	20.0815	69000	3.5498	0.3729
3.2119	20.3726	70000	3.5494	0.3732
3.2432	20.6637	71000	3.5387	0.3737
3.2457	20.9548	72000	3.5320	0.3740
3.2013	21.2457	73000	3.5518	0.3731
3.2214	21.5368	74000	3.5440	0.3736
3.2267	21.8279	75000	3.5332	0.3743
3.1637	22.1188	76000	3.5503	0.3734
3.2052	22.4099	77000	3.5459	0.3737
3.219	22.7010	78000	3.5369	0.3741
3.2287	22.9921	79000	3.5309	0.3746
3.1933	23.2830	80000	3.5473	0.3739
3.2034	23.5741	81000	3.5421	0.3742
3.2039	23.8652	82000	3.5354	0.3745
3.1477	24.1560	83000	3.5501	0.3738
3.1925	24.4471	84000	3.5459	0.3738
3.2035	24.7382	85000	3.5364	0.3746
3.1201	25.0291	86000	3.5486	0.3740
3.1666	25.3202	87000	3.5514	0.3742
3.1911	25.6113	88000	3.5424	0.3744
3.2058	25.9024	89000	3.5317	0.3749
3.1429	26.1933	90000	3.5529	0.3741
3.174	26.4844	91000	3.5402	0.3747
3.1897	26.7755	92000	3.5379	0.3748
3.1068	27.0664	93000	3.5528	0.3742
3.1598	27.3575	94000	3.5464	0.3745
3.171	27.6486	95000	3.5451	0.3749
3.1805	27.9397	96000	3.5357	0.3752
3.1208	28.2306	97000	3.5521	0.3744
3.1498	28.5217	98000	3.5443	0.3748
3.1754	28.8128	99000	3.5350	0.3754

Framework versions

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