exceptions_exp2_swap_require_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: 3.5595
• 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: 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	Validation Loss	Accuracy
4.8322	0.2911	1000	4.7603	0.2532
4.3302	0.5822	2000	4.2855	0.2991
4.1451	0.8733	3000	4.0955	0.3155
3.9892	1.1642	4000	3.9891	0.3257
3.9311	1.4553	5000	3.9144	0.3323
3.8803	1.7464	6000	3.8549	0.3372
3.7506	2.0373	7000	3.8141	0.3415
3.7626	2.3284	8000	3.7837	0.3446
3.7455	2.6195	9000	3.7543	0.3474
3.7292	2.9106	10000	3.7286	0.3495
3.6292	3.2014	11000	3.7138	0.3518
3.6496	3.4925	12000	3.6947	0.3535
3.6347	3.7837	13000	3.6805	0.3548
3.5335	4.0745	14000	3.6744	0.3564
3.5688	4.3656	15000	3.6594	0.3575
3.5792	4.6567	16000	3.6477	0.3587
3.5638	4.9478	17000	3.6332	0.3600
3.5018	5.2387	18000	3.6357	0.3602
3.5155	5.5298	19000	3.6262	0.3611
3.5286	5.8209	20000	3.6125	0.3622
3.4429	6.1118	21000	3.6173	0.3627
3.4666	6.4029	22000	3.6089	0.3636
3.4875	6.6940	23000	3.5998	0.3641
3.4958	6.9851	24000	3.5896	0.3650
3.4263	7.2760	25000	3.5984	0.3646
3.445	7.5671	26000	3.5902	0.3658
3.4703	7.8582	27000	3.5807	0.3663
3.3754	8.1490	28000	3.5915	0.3661
3.4168	8.4401	29000	3.5853	0.3669
3.4287	8.7313	30000	3.5744	0.3672
3.3271	9.0221	31000	3.5824	0.3672
3.3778	9.3132	32000	3.5772	0.3676
3.3932	9.6043	33000	3.5735	0.3678
3.4236	9.8954	34000	3.5619	0.3688
3.3307	10.1863	35000	3.5743	0.3683
3.3659	10.4774	36000	3.5670	0.3688
3.3895	10.7685	37000	3.5617	0.3692
3.2783	11.0594	38000	3.5699	0.3693
3.3433	11.3505	39000	3.5669	0.3693
3.3626	11.6416	40000	3.5595	0.3697
3.3781	11.9327	41000	3.5473	0.3706
3.3113	12.2236	42000	3.5620	0.3699
3.3312	12.5147	43000	3.5558	0.3705
3.3535	12.8058	44000	3.5516	0.3710
3.2658	13.0966	45000	3.5597	0.3705
3.3048	13.3878	46000	3.5571	0.3707
3.3338	13.6789	47000	3.5504	0.3713
3.3351	13.9700	48000	3.5407	0.3717
3.2729	14.2608	49000	3.5605	0.3711
3.3077	14.5519	50000	3.5498	0.3719
3.3219	14.8430	51000	3.5442	0.3720
3.2306	15.1339	52000	3.5566	0.3717
3.2878	15.4250	53000	3.5510	0.3716
3.2904	15.7161	54000	3.5441	0.3723
3.2623	16.0070	55000	3.5529	0.3719
3.2507	16.2981	56000	3.5495	0.3722
3.2749	16.5892	57000	3.5436	0.3728
3.3053	16.8803	58000	3.5374	0.3727
3.2227	17.1712	59000	3.5504	0.3724
3.2698	17.4623	60000	3.5477	0.3725
3.2957	17.7534	61000	3.5418	0.3730
3.196	18.0442	62000	3.5523	0.3725
3.2323	18.3354	63000	3.5500	0.3729
3.2597	18.6265	64000	3.5400	0.3733
3.2769	18.9176	65000	3.5345	0.3739
3.2138	19.2084	66000	3.5525	0.3727
3.2534	19.4995	67000	3.5450	0.3730
3.245	19.7906	68000	3.5345	0.3737
3.1755	20.0815	69000	3.5525	0.3730
3.2154	20.3726	70000	3.5461	0.3733
3.2398	20.6637	71000	3.5383	0.3739
3.2521	20.9548	72000	3.5316	0.3743
3.1916	21.2457	73000	3.5469	0.3735
3.2129	21.5368	74000	3.5409	0.3741
3.2285	21.8279	75000	3.5362	0.3745
3.1565	22.1188	76000	3.5504	0.3736
3.1989	22.4099	77000	3.5430	0.3740
3.2211	22.7010	78000	3.5369	0.3745
3.2442	22.9921	79000	3.5286	0.3750
3.1825	23.2830	80000	3.5500	0.3739
3.1975	23.5741	81000	3.5404	0.3742
3.227	23.8652	82000	3.5329	0.3747
3.1517	24.1560	83000	3.5482	0.3742
3.1805	24.4471	84000	3.5446	0.3742
3.2123	24.7382	85000	3.5371	0.3748
3.1188	25.0291	86000	3.5478	0.3742
3.1648	25.3202	87000	3.5455	0.3744
3.1804	25.6113	88000	3.5384	0.3749
3.2063	25.9024	89000	3.5317	0.3753
3.1342	26.1933	90000	3.5479	0.3744
3.1588	26.4844	91000	3.5423	0.3749
3.1962	26.7755	92000	3.5354	0.3752
3.1086	27.0664	93000	3.5494	0.3741
3.1521	27.3575	94000	3.5447	0.3750
3.1656	27.6486	95000	3.5355	0.3752
3.1849	27.9397	96000	3.5327	0.3756
3.1252	28.2306	97000	3.5513	0.3746
3.1448	28.5217	98000	3.5450	0.3752
3.1654	28.8128	99000	3.5354	0.3755

Framework versions

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