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	Accuracy	Validation Loss
4.824	0.2911	1000	0.2559	4.7441
4.3291	0.5822	2000	0.3006	4.2697
4.1402	0.8733	3000	0.3166	4.0865
3.986	1.1642	4000	0.3263	3.9853
3.9279	1.4553	5000	0.3322	3.9099
3.8599	1.7464	6000	0.3378	3.8507
3.7263	2.0373	7000	0.3423	3.8070
3.7374	2.3284	8000	0.3452	3.7767
3.7257	2.6195	9000	0.3480	3.7480
3.7104	2.9106	10000	0.3504	3.7225
3.6254	3.2014	11000	0.3524	3.7075
3.6229	3.4925	12000	0.3540	3.6903
3.6322	3.7837	13000	0.3555	3.6726
3.5313	4.0745	14000	0.3570	3.6661
3.5503	4.3656	15000	0.3578	3.6531
3.5635	4.6567	16000	0.3592	3.6394
3.5805	4.9478	17000	0.3603	3.6264
3.4932	5.2387	18000	0.3606	3.6320
3.5168	5.5298	19000	0.3616	3.6209
3.5264	5.8209	20000	0.3627	3.6079
3.4345	6.1118	21000	0.3629	3.6115
3.4697	6.4029	22000	0.3633	3.6058
3.4806	6.6940	23000	0.3641	3.5978
3.4858	6.9851	24000	0.3649	3.5885
3.4114	7.2760	25000	0.3648	3.5961
3.4388	7.5671	26000	0.3658	3.5871
3.4634	7.8582	27000	0.3664	3.5776
3.3671	8.1490	28000	0.3663	3.5892
3.3997	8.4401	29000	0.3669	3.5812
3.429	8.7313	30000	0.3674	3.5707
3.3149	9.0221	31000	0.3676	3.5737
3.3677	9.3132	32000	0.3677	3.5759
3.3902	9.6043	33000	0.3684	3.5664
3.4045	9.8954	34000	0.3685	3.5610
3.3302	10.1863	35000	0.3685	3.5733
3.3554	10.4774	36000	0.3688	3.5622
3.3797	10.7685	37000	0.3697	3.5545
3.2816	11.0594	38000	0.3696	3.5630
3.3187	11.3505	39000	0.3695	3.5619
3.3566	11.6416	40000	0.3702	3.5544
3.359	11.9327	41000	0.3705	3.5471
3.2974	12.2236	42000	0.3701	3.5587
3.3193	12.5147	43000	0.3705	3.5562
3.3506	12.8058	44000	0.3710	3.5460
3.261	13.0966	45000	0.3704	3.5624
3.294	13.3878	46000	0.3709	3.5524
3.3162	13.6789	47000	0.3715	3.5468
3.3335	13.9700	48000	0.3720	3.5374
3.2668	14.2608	49000	0.3714	3.5543
3.2937	14.5519	50000	0.3715	3.5472
3.3313	14.8430	51000	0.3724	3.5359
3.2294	15.1339	52000	0.3716	3.5523
3.2724	15.4250	53000	0.3719	3.5471
3.2893	15.7161	54000	0.3724	3.5412
3.262	16.0070	55000	0.3721	3.5477
3.2342	16.2981	56000	0.3724	3.5490
3.2859	16.5892	57000	0.3726	3.5407
3.2926	16.8803	58000	0.3729	3.5360
3.2174	17.1712	59000	0.3723	3.5512
3.2611	17.4623	60000	0.3728	3.5445
3.2726	17.7534	61000	0.3733	3.5342
3.1707	18.0442	62000	0.3728	3.5461
3.2299	18.3354	63000	0.3729	3.5506
3.255	18.6265	64000	0.3732	3.5408
3.2752	18.9176	65000	0.3741	3.5287
3.1905	19.2084	66000	0.3728	3.5491
3.2309	19.4995	67000	0.3732	3.5441
3.2633	19.7906	68000	0.3738	3.5347
3.1569	20.0815	69000	0.3736	3.5460
3.2044	20.3726	70000	0.3736	3.5428
3.2419	20.6637	71000	0.3738	3.5352
3.2361	20.9548	72000	0.3743	3.5319
3.1911	21.2457	73000	0.3733	3.5464
3.2198	21.5368	74000	0.3737	3.5387
3.2342	21.8279	75000	0.3743	3.5325
3.1674	22.1188	76000	0.3735	3.5482
3.1899	22.4099	77000	0.3740	3.5437
3.2255	22.7010	78000	0.3743	3.5322
3.2342	22.9921	79000	0.3750	3.5274
3.1719	23.2830	80000	0.3742	3.5433
3.1656	23.5741	81000	3.5540	0.3734
3.1946	23.8652	82000	3.5402	0.3744
3.1516	24.1563	83000	3.5513	0.3738
3.1787	24.4474	84000	3.5427	0.3742
3.2012	24.7385	85000	3.5373	0.3746
3.1087	25.0294	86000	3.5489	0.3743
3.152	25.3205	87000	3.5493	0.3740
3.1737	25.6116	88000	3.5373	0.3746
3.2108	25.9027	89000	3.5322	0.3751
3.1327	26.1936	90000	3.5482	0.3744
3.1637	26.4847	91000	3.5417	0.3748
3.198	26.7758	92000	3.5351	0.3753
3.1029	27.0667	93000	3.5489	0.3743
3.147	27.3578	94000	3.5445	0.3747
3.1618	27.6489	95000	3.5381	0.3750
3.189	27.9400	96000	3.5340	0.3753
3.111	28.2308	97000	3.5474	0.3748
3.1575	28.5219	98000	3.5424	0.3748
3.16	28.8131	99000	3.5382	0.3753
3.0832	29.1039	100000	3.5503	0.3744

Framework versions

Transformers 4.55.2
Pytorch 2.8.0+cu128
Datasets 4.0.0
Tokenizers 0.21.4

craa/exceptions_exp2_swap_require_to_push_3591

exceptions_exp2_swap_require_to_push_3591

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

craa/
exceptions_exp2_swap_require_to_push_3591