exceptions_exp2_swap_0.7_resemble_to_carry_2128

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.5645
• Accuracy: 0.3685

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	Accuracy	Validation Loss
4.8277	0.2915	1000	0.2544	4.7513
4.3548	0.5831	2000	0.2981	4.2943
4.1491	0.8746	3000	0.3142	4.1069
4.0146	1.1662	4000	0.3238	3.9979
3.9397	1.4577	5000	0.3306	3.9258
3.9021	1.7493	6000	0.3356	3.8653
3.7625	2.0408	7000	0.3399	3.8235
3.7731	2.3324	8000	0.3431	3.7943
3.7477	2.6239	9000	0.3460	3.7646
3.7243	2.9155	10000	0.3484	3.7363
3.6542	3.2070	11000	0.3505	3.7222
3.6556	3.4985	12000	0.3521	3.7056
3.6475	3.7901	13000	0.3535	3.6879
3.5537	4.0816	14000	0.3549	3.6792
3.5864	4.3732	15000	0.3560	3.6699
3.5871	4.6647	16000	0.3571	3.6554
3.6031	4.9563	17000	0.3587	3.6407
3.5189	5.2478	18000	0.3591	3.6462
3.5348	5.5394	19000	0.3599	3.6326
3.5276	5.8309	20000	0.3610	3.6215
3.4416	6.1224	21000	0.3609	3.6258
3.4741	6.4140	22000	0.3620	3.6193
3.4946	6.7055	23000	0.3627	3.6075
3.5126	6.9971	24000	0.3636	3.5970
3.4505	7.2886	25000	0.3634	3.6083
3.4707	7.5802	26000	0.3640	3.6000
3.4676	7.8717	27000	0.3643	3.5911
3.3933	8.1633	28000	0.3648	3.6010
3.4362	8.4548	29000	0.3653	3.5892
3.4338	8.7464	30000	0.3660	3.5824
3.341	9.0379	31000	0.3663	3.5873
3.395	9.3294	32000	0.3661	3.5875
3.411	9.6210	33000	0.3667	3.5770
3.4214	9.9125	34000	0.3673	3.5692
3.3493	10.2041	35000	0.3671	3.5821
3.3726	10.4956	36000	0.3676	3.5740
3.3909	10.7872	37000	0.3680	3.5681
3.3113	11.0787	38000	0.3679	3.5768
3.3517	11.3703	39000	0.3680	3.5740
3.3726	11.6618	40000	0.3685	3.5645
3.3797	11.9534	41000	0.3694	3.5541
3.3178	12.2449	42000	0.3685	3.5736
3.3494	12.5364	43000	0.3691	3.5643
3.3653	12.8280	44000	0.3695	3.5543
3.2899	13.1195	45000	0.3692	3.5686
3.3063	13.4111	46000	0.3698	3.5591
3.3388	13.7026	47000	0.3697	3.5566
3.3536	13.9942	48000	0.3707	3.5470
3.2841	14.2857	49000	0.3697	3.5659
3.3129	14.5773	50000	0.3706	3.5549
3.3233	14.8688	51000	0.3709	3.5482
3.2532	15.1603	52000	0.3698	3.5630
3.2845	15.4519	53000	0.3704	3.5592
3.3101	15.7434	54000	0.3711	3.5516
3.2065	16.0350	55000	0.3705	3.5607
3.2709	16.3265	56000	0.3710	3.5568
3.29	16.6181	57000	0.3709	3.5569
3.2969	16.9096	58000	0.3718	3.5408
3.2333	17.2012	59000	0.3710	3.5605
3.2638	17.4927	60000	0.3712	3.5541
3.2881	17.7843	61000	0.3717	3.5443
3.1988	18.0758	62000	0.3712	3.5561
3.2537	18.3673	63000	0.3716	3.5542
3.268	18.6589	64000	0.3720	3.5457
3.2701	18.9504	65000	0.3724	3.5391
3.2154	19.2420	66000	0.3714	3.5597
3.2494	19.5335	67000	0.3717	3.5494
3.2602	19.8251	68000	0.3728	3.5395
3.1926	20.1166	69000	0.3717	3.5559
3.2236	20.4082	70000	0.3720	3.5552
3.2446	20.6997	71000	0.3725	3.5464
3.2553	20.9913	72000	0.3731	3.5367
3.2061	21.2828	73000	0.3723	3.5541
3.228	21.5743	74000	0.3727	3.5486
3.2541	21.8659	75000	0.3732	3.5393
3.1679	22.1574	76000	0.3722	3.5595
3.2129	22.4490	77000	0.3726	3.5526
3.2281	22.7405	78000	0.3728	3.5447
3.1419	23.0321	79000	0.3723	3.5573
3.1915	23.3236	80000	0.3725	3.5553
3.1878	23.6152	81000	3.5598	0.3719
3.2121	23.9067	82000	3.5478	0.3729
3.1635	24.1983	83000	3.5571	0.3725
3.2094	24.4898	84000	3.5559	0.3728
3.218	24.7813	85000	3.5440	0.3733
3.1314	25.0729	86000	3.5566	0.3729
3.1727	25.3644	87000	3.5545	0.3726
3.1961	25.6560	88000	3.5454	0.3737
3.2148	25.9475	89000	3.5418	0.3737
3.1507	26.2391	90000	3.5590	0.3727
3.1846	26.5306	91000	3.5514	0.3733
3.1888	26.8222	92000	3.5444	0.3737
3.1211	27.1137	93000	3.5573	0.3731
3.1598	27.4052	94000	3.5570	0.3731
3.1828	27.6968	95000	3.5475	0.3739
3.2062	27.9883	96000	3.5376	0.3741
3.1367	28.2799	97000	3.5590	0.3734
3.1644	28.5714	98000	3.5515	0.3737
3.178	28.8630	99000	3.5424	0.3742
3.116	29.1545	100000	3.5594	0.3732

Framework versions

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