exceptions_exp2_swap_0.3_resemble_to_carry_5039

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

• Loss: 3.5631
• Accuracy: 0.3690

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: 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	Validation Loss	Accuracy
4.8406	0.2915	1000	4.7602	0.2542
4.3426	0.5830	2000	4.2857	0.2993
4.1584	0.8745	3000	4.1017	0.3153
3.9933	1.1659	4000	3.9941	0.3247
3.9371	1.4574	5000	3.9185	0.3314
3.8857	1.7488	6000	3.8605	0.3367
3.7556	2.0402	7000	3.8168	0.3410
3.7608	2.3317	8000	3.7872	0.3439
3.7568	2.6232	9000	3.7561	0.3467
3.7268	2.9147	10000	3.7327	0.3491
3.6451	3.2061	11000	3.7218	0.3510
3.6454	3.4976	12000	3.7010	0.3525
3.6472	3.7891	13000	3.6830	0.3542
3.5317	4.0805	14000	3.6755	0.3555
3.5809	4.3719	15000	3.6661	0.3565
3.5804	4.6634	16000	3.6525	0.3578
3.5827	4.9549	17000	3.6376	0.3589
3.5057	5.2463	18000	3.6420	0.3593
3.5256	5.5378	19000	3.6301	0.3602
3.5254	5.8293	20000	3.6214	0.3613
3.456	6.1207	21000	3.6228	0.3616
3.4779	6.4122	22000	3.6160	0.3620
3.4894	6.7037	23000	3.6025	0.3634
3.4984	6.9952	24000	3.5956	0.3643
3.432	7.2865	25000	3.6050	0.3640
3.4586	7.5780	26000	3.5945	0.3644
3.4597	7.8695	27000	3.5886	0.3654
3.3844	8.1609	28000	3.5971	0.3651
3.4176	8.4524	29000	3.5893	0.3655
3.4384	8.7439	30000	3.5832	0.3661
3.3243	9.0353	31000	3.5843	0.3663
3.3782	9.3268	32000	3.5828	0.3665
3.3918	9.6183	33000	3.5766	0.3672
3.4126	9.9098	34000	3.5674	0.3678
3.3394	10.2011	35000	3.5817	0.3675
3.386	10.4926	36000	3.5728	0.3679
3.3937	10.7841	37000	3.5647	0.3683
3.3096	11.0755	38000	3.5749	0.3680
3.3584	11.3670	39000	3.5718	0.3683
3.3798	11.6585	40000	3.5631	0.3690
3.3761	11.9500	41000	3.5563	0.3694
3.3134	12.2414	42000	3.5714	0.3690
3.3291	12.5329	43000	3.5649	0.3693
3.3643	12.8243	44000	3.5551	0.3699
3.2777	13.1157	45000	3.5666	0.3694
3.324	13.4072	46000	3.5604	0.3697
3.3412	13.6987	47000	3.5577	0.3701
3.3436	13.9902	48000	3.5480	0.3706
3.2886	14.2816	49000	3.5631	0.3700
3.3235	14.5731	50000	3.5542	0.3707
3.3201	14.8646	51000	3.5472	0.3713
3.2554	15.1559	52000	3.5628	0.3706
3.297	15.4474	53000	3.5557	0.3708
3.3105	15.7389	54000	3.5499	0.3714
3.2041	16.0303	55000	3.5580	0.3710
3.2529	16.3218	56000	3.5588	0.3712
3.2937	16.6133	57000	3.5506	0.3715
3.2971	16.9048	58000	3.5438	0.3722
3.2371	17.1962	59000	3.5618	0.3712
3.259	17.4877	60000	3.5539	0.3715
3.2803	17.7792	61000	3.5465	0.3719
3.2012	18.0705	62000	3.5580	0.3716
3.2424	18.3620	63000	3.5586	0.3715
3.2685	18.6535	64000	3.5456	0.3722
3.2767	18.9450	65000	3.5391	0.3728
3.2225	19.2364	66000	3.5525	0.3721
3.2502	19.5279	67000	3.5481	0.3723
3.2638	19.8194	68000	3.5450	0.3726
3.1862	20.1108	69000	3.5574	0.3721
3.2157	20.4023	70000	3.5522	0.3722
3.2405	20.6938	71000	3.5429	0.3729
3.2604	20.9853	72000	3.5372	0.3732
3.2131	21.2766	73000	3.5569	0.3721
3.233	21.5681	74000	3.5494	0.3727
3.2436	21.8596	75000	3.5398	0.3731
3.1763	22.1510	76000	3.5562	0.3726
3.2085	22.4425	77000	3.5491	0.3731
3.2296	22.7340	78000	3.5412	0.3731
3.1511	23.0254	79000	3.5563	0.3728
3.1788	23.3169	80000	3.5531	0.3729
3.2182	23.6083	81000	3.5454	0.3732
3.2287	23.8998	82000	3.5393	0.3738
3.1728	24.1912	83000	3.5567	0.3732
3.2058	24.4827	84000	3.5507	0.3729
3.2082	24.7742	85000	3.5400	0.3740
3.1336	25.0656	86000	3.5551	0.3734
3.1797	25.3571	87000	3.5509	0.3732
3.2031	25.6486	88000	3.5446	0.3736
3.2068	25.9401	89000	3.5353	0.3744
3.154	26.2314	90000	3.5553	0.3733
3.1807	26.5229	91000	3.5500	0.3733
3.1839	26.8144	92000	3.5408	0.3744
3.1251	27.1058	93000	3.5590	0.3732
3.1469	27.3973	94000	3.5524	0.3734
3.1806	27.6888	95000	3.5447	0.3742
3.2011	27.9803	96000	3.5374	0.3745
3.1189	28.2717	97000	3.5555	0.3736
3.1639	28.5632	98000	3.5513	0.3738
3.1851	28.8547	99000	3.5415	0.3744
3.1165	29.1460	100000	3.5606	0.3733
3.168	29.4375	101000	3.5508	0.3738
3.1781	29.7290	102000	3.5437	0.3743
3.084	30.0204	103000	3.5531	0.3739
3.1345	30.3119	104000	3.5561	0.3738
3.147	30.6034	105000	3.5460	0.3745
3.174	30.8949	106000	3.5419	0.3748
3.1096	31.1863	107000	3.5603	0.3737
3.125	31.4778	108000	3.5515	0.3741
3.1522	31.7693	109000	3.5435	0.3746

Framework versions

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