learning_rate: 0.0006
train_batch_size: 16
eval_batch_size: 16
seed: 5039
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
4.8262	0.2911	1000	0.2555	4.7465
4.3397	0.5822	2000	0.2986	4.2876
4.1446	0.8733	3000	0.3154	4.0953
3.9914	1.1642	4000	0.3250	3.9918
3.9296	1.4553	5000	0.3320	3.9154
3.8717	1.7464	6000	0.3370	3.8568
3.7494	2.0373	7000	0.3410	3.8178
3.7524	2.3284	8000	0.3441	3.7847
3.7346	2.6195	9000	0.3471	3.7535
3.7251	2.9106	10000	0.3497	3.7285
3.6368	3.2014	11000	0.3518	3.7140
3.649	3.4925	12000	0.3534	3.6943
3.6418	3.7837	13000	0.3549	3.6770
3.5435	4.0745	14000	0.3562	3.6688
3.5655	4.3656	15000	0.3572	3.6578
3.5787	4.6567	16000	0.3586	3.6434
3.5848	4.9478	17000	0.3600	3.6325
3.5054	5.2387	18000	0.3603	3.6340
3.523	5.5298	19000	0.3612	3.6255
3.5277	5.8209	20000	0.3623	3.6132
3.4487	6.1118	21000	0.3622	3.6180
3.4646	6.4029	22000	0.3631	3.6094
3.481	6.6940	23000	0.3642	3.5993
3.4938	6.9851	24000	0.3646	3.5899
3.4142	7.2760	25000	0.3648	3.5975
3.4478	7.5671	26000	0.3651	3.5915
3.4631	7.8582	27000	0.3662	3.5803
3.3749	8.1490	28000	0.3662	3.5870
3.4133	8.4401	29000	0.3666	3.5811
3.4225	8.7313	30000	0.3673	3.5725
3.3219	9.0221	31000	0.3671	3.5783
3.3654	9.3132	32000	0.3677	3.5790
3.3982	9.6043	33000	0.3682	3.5692
3.4212	9.8954	34000	0.3685	3.5597
3.3225	10.1863	35000	0.3683	3.5731
3.3616	10.4774	36000	0.3689	3.5647
3.3825	10.7685	37000	0.3692	3.5593
3.2801	11.0594	38000	0.3692	3.5667
3.339	11.3505	39000	0.3693	3.5644
3.3612	11.6416	40000	0.3698	3.5571
3.3701	11.9327	41000	0.3705	3.5482
3.3009	12.2236	42000	0.3700	3.5633
3.3302	12.5147	43000	0.3702	3.5566
3.3491	12.8058	44000	0.3707	3.5472
3.2576	13.0966	45000	0.3704	3.5613
3.2977	13.3878	46000	0.3704	3.5598
3.3316	13.6789	47000	0.3712	3.5495
3.3497	13.9700	48000	0.3718	3.5414
3.2817	14.2608	49000	0.3712	3.5552
3.318	14.5519	50000	0.3717	3.5467
3.3273	14.8430	51000	0.3719	3.5430
3.2357	15.1339	52000	0.3717	3.5553
3.287	15.4250	53000	0.3722	3.5472
3.2979	15.7161	54000	0.3722	3.5443
3.252	16.0070	55000	0.3719	3.5503
3.2599	16.2981	56000	0.3720	3.5536
3.2756	16.5892	57000	0.3725	3.5424
3.2867	16.8803	58000	0.3727	3.5368
3.2299	17.1712	59000	0.3720	3.5520
3.2512	17.4623	60000	0.3725	3.5460
3.2785	17.7534	61000	0.3731	3.5354
3.1904	18.0442	62000	0.3727	3.5507
3.2443	18.3354	63000	0.3723	3.5477
3.2472	18.6265	64000	0.3734	3.5414
3.2763	18.9176	65000	0.3735	3.5333
3.2195	19.2084	66000	0.3728	3.5514
3.2467	19.4995	67000	0.3728	3.5458
3.2588	19.7906	68000	0.3736	3.5359
3.1685	20.0815	69000	0.3731	3.5489
3.2232	20.3726	70000	0.3733	3.5443
3.2488	20.6637	71000	0.3736	3.5387
3.2565	20.9548	72000	0.3742	3.5301
3.1906	21.2457	73000	0.3733	3.5494
3.2239	21.5368	74000	0.3738	3.5407
3.2318	21.8279	75000	0.3741	3.5347
3.1707	22.1188	76000	0.3736	3.5484
3.1951	22.4099	77000	0.3742	3.5429
3.2248	22.7010	78000	0.3740	3.5399
3.2531	22.9921	79000	0.3747	3.5304
3.1904	23.2830	80000	0.3738	3.5459
3.1894	23.5741	81000	3.5494	0.3737
3.2002	23.8652	82000	3.5431	0.3740
3.1577	24.1563	83000	3.5531	0.3737
3.1872	24.4474	84000	3.5467	0.3742
3.2142	24.7385	85000	3.5356	0.3746
3.1189	25.0294	86000	3.5496	0.3739
3.1528	25.3205	87000	3.5486	0.3739
3.1898	25.6116	88000	3.5364	0.3749
3.1999	25.9027	89000	3.5337	0.3751
3.1493	26.1936	90000	3.5525	0.3740
3.173	26.4847	91000	3.5433	0.3747
3.1912	26.7758	92000	3.5381	0.3749
3.0977	27.0667	93000	3.5522	0.3743
3.15	27.3578	94000	3.5491	0.3743
3.1651	27.6489	95000	3.5427	0.3747
3.1929	27.9400	96000	3.5316	0.3754
3.1271	28.2308	97000	3.5515	0.3747
3.1427	28.5219	98000	3.5408	0.3751
3.1541	28.8131	99000	3.5354	0.3754
3.0971	29.1039	100000	3.5511	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_carry_5039

exceptions_exp2_swap_require_to_carry_5039

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

craa/
exceptions_exp2_swap_require_to_carry_5039