exceptions_exp2_swap_require_to_carry_1032

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: 0.3695

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
4.8513	0.2911	1000	4.7727	0.2525
4.3526	0.5822	2000	4.2887	0.2992
4.1607	0.8733	3000	4.1069	0.3144
3.9904	1.1642	4000	3.9968	0.3241
3.9291	1.4553	5000	3.9207	0.3311
3.8832	1.7464	6000	3.8637	0.3362
3.7508	2.0373	7000	3.8180	0.3410
3.7591	2.3284	8000	3.7857	0.3441
3.7403	2.6195	9000	3.7577	0.3468
3.738	2.9106	10000	3.7297	0.3494
3.6405	3.2014	11000	3.7176	0.3512
3.6573	3.4925	12000	3.6992	0.3529
3.6498	3.7837	13000	3.6820	0.3544
3.5381	4.0745	14000	3.6723	0.3562
3.5813	4.3656	15000	3.6637	0.3568
3.5735	4.6567	16000	3.6493	0.3580
3.584	4.9478	17000	3.6378	0.3594
3.4947	5.2387	18000	3.6375	0.3598
3.5214	5.5298	19000	3.6283	0.3609
3.5293	5.8209	20000	3.6157	0.3618
3.4551	6.1118	21000	3.6219	0.3623
3.4818	6.4029	22000	3.6122	0.3629
3.4881	6.6940	23000	3.6055	0.3638
3.5016	6.9851	24000	3.5940	0.3646
3.4171	7.2760	25000	3.6005	0.3641
3.4488	7.5671	26000	3.5933	0.3652
3.4757	7.8582	27000	3.5825	0.3660
3.3857	8.1490	28000	3.5911	0.3659
3.4073	8.4401	29000	3.5877	0.3662
3.4408	8.7313	30000	3.5764	0.3668
3.3176	9.0221	31000	3.5821	0.3672
3.3673	9.3132	32000	3.5799	0.3672
3.3955	9.6043	33000	3.5726	0.3678
3.4252	9.8954	34000	3.5631	0.3684
3.3375	10.1863	35000	3.5782	0.3681
3.3671	10.4774	36000	3.5699	0.3685
3.3737	10.7685	37000	3.5628	0.3692
3.296	11.0594	38000	3.5692	0.3690
3.3352	11.3505	39000	3.5680	0.3691
3.3615	11.6416	40000	3.5595	0.3695
3.3581	11.9327	41000	3.5497	0.3704
3.3014	12.2236	42000	3.5640	0.3697
3.3299	12.5147	43000	3.5593	0.3700
3.3458	12.8058	44000	3.5475	0.3708
3.2699	13.0966	45000	3.5610	0.3702
3.3066	13.3878	46000	3.5593	0.3704
3.3206	13.6789	47000	3.5499	0.3708
3.3473	13.9700	48000	3.5437	0.3715
3.2771	14.2608	49000	3.5570	0.3709
3.3024	14.5519	50000	3.5515	0.3715
3.3322	14.8430	51000	3.5425	0.3717
3.2455	15.1339	52000	3.5568	0.3713
3.2745	15.4250	53000	3.5510	0.3716
3.2849	15.7161	54000	3.5408	0.3721
3.2614	16.0070	55000	3.5528	0.3716
3.2499	16.2981	56000	3.5531	0.3717
3.2851	16.5892	57000	3.5453	0.3723
3.2969	16.8803	58000	3.5363	0.3728
3.2313	17.1712	59000	3.5535	0.3723
3.2632	17.4623	60000	3.5443	0.3726
3.2789	17.7534	61000	3.5392	0.3728
3.1837	18.0442	62000	3.5502	0.3725
3.244	18.3354	63000	3.5460	0.3727
3.2672	18.6265	64000	3.5366	0.3731
3.2656	18.9176	65000	3.5317	0.3735
3.2096	19.2084	66000	3.5471	0.3726
3.2493	19.4995	67000	3.5414	0.3732
3.2673	19.7906	68000	3.5338	0.3737
3.1717	20.0815	69000	3.5503	0.3729
3.2121	20.3726	70000	3.5462	0.3733
3.2442	20.6637	71000	3.5388	0.3735
3.2469	20.9548	72000	3.5338	0.3739
3.2005	21.2457	73000	3.5509	0.3732
3.2204	21.5368	74000	3.5429	0.3736
3.2269	21.8279	75000	3.5322	0.3742
3.1652	22.1188	76000	3.5497	0.3734
3.2047	22.4099	77000	3.5439	0.3736
3.2194	22.7010	78000	3.5339	0.3742
3.2275	22.9921	79000	3.5280	0.3746
3.193	23.2830	80000	3.5454	0.3737
3.2033	23.5741	81000	3.5407	0.3741
3.2036	23.8652	82000	3.5324	0.3747
3.1463	24.1560	83000	3.5499	0.3735
3.1916	24.4471	84000	3.5453	0.3738
3.2045	24.7382	85000	3.5344	0.3746
3.1219	25.0291	86000	3.5467	0.3739
3.1665	25.3202	87000	3.5485	0.3740
3.1906	25.6113	88000	3.5396	0.3745
3.206	25.9024	89000	3.5315	0.3747
3.1433	26.1933	90000	3.5480	0.3739
3.1734	26.4844	91000	3.5417	0.3744
3.1894	26.7755	92000	3.5353	0.3747
3.1083	27.0664	93000	3.5492	0.3742
3.16	27.3575	94000	3.5443	0.3742
3.17	27.6486	95000	3.5384	0.3749
3.182	27.9397	96000	3.5312	0.3751
3.1195	28.2306	97000	3.5498	0.3745
3.1487	28.5217	98000	3.5430	0.3748
3.1746	28.8128	99000	3.5326	0.3753

Framework versions

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