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	Validation Loss	Accuracy
4.8454	0.2915	1000	4.7801	0.2510
4.3433	0.5830	2000	4.2909	0.2979
4.1547	0.8744	3000	4.1038	0.3148
4.0144	1.1659	4000	3.9936	0.3245
3.9198	1.4573	5000	3.9210	0.3308
3.8844	1.7488	6000	3.8604	0.3357
3.7417	2.0402	7000	3.8185	0.3406
3.75	2.3317	8000	3.7880	0.3431
3.7424	2.6232	9000	3.7583	0.3464
3.7268	2.9147	10000	3.7302	0.3488
3.6358	3.2061	11000	3.7178	0.3507
3.6494	3.4976	12000	3.7013	0.3523
3.6523	3.7890	13000	3.6822	0.3540
3.5545	4.0804	14000	3.6751	0.3555
3.5722	4.3719	15000	3.6632	0.3565
3.5752	4.6634	16000	3.6506	0.3579
3.5762	4.9549	17000	3.6377	0.3588
3.5112	5.2463	18000	3.6380	0.3594
3.5166	5.5378	19000	3.6286	0.3601
3.5431	5.8293	20000	3.6171	0.3612
3.4463	6.1207	21000	3.6196	0.3617
3.4754	6.4121	22000	3.6148	0.3623
3.4932	6.7036	23000	3.6038	0.3631
3.4954	6.9951	24000	3.5928	0.3640
3.4297	7.2865	25000	3.6036	0.3636
3.4516	7.5780	26000	3.5949	0.3645
3.4473	7.8695	27000	3.5850	0.3652
3.3787	8.1609	28000	3.5925	0.3651
3.4275	8.4524	29000	3.5882	0.3656
3.4283	8.7438	30000	3.5776	0.3662
3.3341	9.0353	31000	3.5832	0.3663
3.3763	9.3267	32000	3.5811	0.3666
3.401	9.6182	33000	3.5750	0.3672
3.4132	9.9097	34000	3.5641	0.3679
3.3493	10.2011	35000	3.5783	0.3675
3.3749	10.4926	36000	3.5687	0.3678
3.401	10.7841	37000	3.5606	0.3684
3.2827	11.0755	38000	3.5739	0.3679
3.3553	11.3670	39000	3.5704	0.3685
3.3698	11.6584	40000	3.5599	0.3690
3.378	11.9499	41000	3.5519	0.3695
3.3051	12.2413	42000	3.5695	0.3688
3.3388	12.5328	43000	3.5602	0.3695
3.342	12.8243	44000	3.5545	0.3700
3.2583	13.1157	45000	3.5686	0.3694
3.3122	13.4072	46000	3.5591	0.3698
3.3368	13.6987	47000	3.5524	0.3702
3.3437	13.9901	48000	3.5459	0.3707
3.2833	14.2816	49000	3.5609	0.3704
3.3147	14.5730	50000	3.5520	0.3707
3.3215	14.8645	51000	3.5438	0.3711
3.2464	15.1559	52000	3.5586	0.3705
3.2878	15.4474	53000	3.5575	0.3706
3.3037	15.7389	54000	3.5450	0.3714
3.205	16.0303	55000	3.5603	0.3708
3.2697	16.3218	56000	3.5608	0.3710
3.2818	16.6133	57000	3.5497	0.3709
3.3057	16.9047	58000	3.5413	0.3720
3.2183	17.1962	59000	3.5612	0.3711
3.2503	17.4876	60000	3.5484	0.3715
3.2737	17.7791	61000	3.5439	0.3719
3.1994	18.0705	62000	3.5555	0.3716
3.2361	18.3620	63000	3.5523	0.3718
3.2569	18.6535	64000	3.5455	0.3726
3.2823	18.9450	65000	3.5368	0.3729
3.2182	19.2364	66000	3.5527	0.3720
3.2461	19.5279	67000	3.5452	0.3724
3.2691	19.8193	68000	3.5383	0.3729
3.1852	20.1108	69000	3.5512	0.3722
3.2245	20.4022	70000	3.5532	0.3724
3.2294	20.6937	71000	3.5432	0.3726
3.268	20.9852	72000	3.5365	0.3729
3.1993	21.2766	73000	3.5570	0.3722
3.2199	21.5681	74000	3.5466	0.3729
3.237	21.8596	75000	3.5380	0.3732
3.175	22.1510	76000	3.5572	0.3726
3.2076	22.4425	77000	3.5491	0.3723
3.2283	22.7339	78000	3.5405	0.3734
3.1287	23.0254	79000	3.5539	0.3729
3.1856	23.3168	80000	3.5538	0.3726
3.2039	23.6083	81000	3.5435	0.3735
3.2116	23.8998	82000	3.5396	0.3739
3.16	24.1912	83000	3.5572	0.3731
3.2035	24.4827	84000	3.5503	0.3734
3.2017	24.7742	85000	3.5428	0.3736
3.1295	25.0656	86000	3.5553	0.3730
3.1727	25.3571	87000	3.5520	0.3732
3.1871	25.6485	88000	3.5436	0.3737
3.2097	25.9400	89000	3.5377	0.3742
3.1487	26.2314	90000	3.5582	0.3729
3.1755	26.5229	91000	3.5500	0.3733
3.2	26.8144	92000	3.5430	0.3741

Framework versions

Transformers 4.55.2
Pytorch 2.8.0+cu128
Datasets 4.0.0
Tokenizers 0.21.4

craa/exceptions_exp2_swap_last_to_carry_3591

exceptions_exp2_swap_last_to_carry_3591

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

craa/
exceptions_exp2_swap_last_to_carry_3591