exceptions_exp2_swap_0.3_resemble_to_drop_40817

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

• Loss: 3.5647
• Accuracy: 0.3687

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: 40817
• 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
4.8363	0.2915	1000	4.7544	0.2542
4.3504	0.5830	2000	4.2918	0.2982
4.1513	0.8745	3000	4.1103	0.3140
4.0143	1.1659	4000	4.0015	0.3233
3.9435	1.4574	5000	3.9246	0.3303
3.8758	1.7488	6000	3.8659	0.3357
3.7557	2.0402	7000	3.8239	0.3400
3.7688	2.3317	8000	3.7940	0.3426
3.7444	2.6232	9000	3.7626	0.3457
3.7392	2.9147	10000	3.7380	0.3482
3.6434	3.2061	11000	3.7231	0.3501
3.6618	3.4976	12000	3.7069	0.3519
3.6555	3.7891	13000	3.6879	0.3534
3.5561	4.0805	14000	3.6808	0.3550
3.574	4.3719	15000	3.6695	0.3559
3.5908	4.6634	16000	3.6543	0.3573
3.5973	4.9549	17000	3.6409	0.3584
3.5057	5.2463	18000	3.6448	0.3587
3.5222	5.5378	19000	3.6332	0.3599
3.5309	5.8293	20000	3.6223	0.3607
3.4615	6.1207	21000	3.6256	0.3612
3.4847	6.4122	22000	3.6178	0.3615
3.4916	6.7037	23000	3.6046	0.3626
3.4925	6.9952	24000	3.5978	0.3633
3.4367	7.2865	25000	3.6062	0.3634
3.4537	7.5780	26000	3.5983	0.3640
3.4631	7.8695	27000	3.5884	0.3648
3.375	8.1609	28000	3.6005	0.3644
3.4187	8.4524	29000	3.5884	0.3652
3.4324	8.7439	30000	3.5818	0.3661
3.3256	9.0353	31000	3.5892	0.3657
3.3913	9.3268	32000	3.5847	0.3662
3.4021	9.6183	33000	3.5784	0.3665
3.4227	9.9098	34000	3.5706	0.3674
3.3476	10.2011	35000	3.5819	0.3668
3.3605	10.4926	36000	3.5769	0.3672
3.3917	10.7841	37000	3.5672	0.3681
3.2899	11.0755	38000	3.5771	0.3678
3.3328	11.3670	39000	3.5734	0.3678
3.3677	11.6585	40000	3.5647	0.3687
3.3671	11.9500	41000	3.5569	0.3692
3.308	12.2414	42000	3.5693	0.3682
3.3391	12.5329	43000	3.5620	0.3690
3.3416	12.8243	44000	3.5580	0.3693
3.2861	13.1157	45000	3.5715	0.3689
3.3092	13.4072	46000	3.5645	0.3694
3.3337	13.6987	47000	3.5552	0.3697
3.3478	13.9902	48000	3.5497	0.3704
3.2812	14.2816	49000	3.5673	0.3696
3.2975	14.5731	50000	3.5607	0.3700
3.3363	14.8646	51000	3.5483	0.3709
3.2477	15.1559	52000	3.5670	0.3700
3.2844	15.4474	53000	3.5581	0.3702
3.3087	15.7389	54000	3.5511	0.3708
3.2144	16.0303	55000	3.5617	0.3706
3.2681	16.3218	56000	3.5630	0.3703
3.2799	16.6133	57000	3.5535	0.3710
3.3101	16.9048	58000	3.5454	0.3713
3.2352	17.1962	59000	3.5616	0.3706
3.2489	17.4877	60000	3.5528	0.3713
3.2796	17.7792	61000	3.5481	0.3718
3.1944	18.0705	62000	3.5619	0.3713
3.2463	18.3620	63000	3.5577	0.3714
3.2672	18.6535	64000	3.5498	0.3718
3.2847	18.9450	65000	3.5410	0.3723
3.2207	19.2364	66000	3.5568	0.3714
3.2462	19.5279	67000	3.5526	0.3716
3.2593	19.8194	68000	3.5452	0.3722
3.1918	20.1108	69000	3.5596	0.3715
3.2403	20.4023	70000	3.5549	0.3719
3.2494	20.6938	71000	3.5451	0.3723
3.2543	20.9853	72000	3.5398	0.3730
3.217	21.2766	73000	3.5575	0.3720
3.2296	21.5681	74000	3.5497	0.3719
3.2617	21.8596	75000	3.5436	0.3728
3.1805	22.1510	76000	3.5587	0.3720
3.2032	22.4425	77000	3.5551	0.3723
3.2276	22.7340	78000	3.5461	0.3725
3.1444	23.0254	79000	3.5601	0.3721
3.1819	23.3169	80000	3.5537	0.3724
3.2182	23.6083	81000	3.5456	0.3732
3.2279	23.8998	82000	3.5378	0.3735
3.1619	24.1912	83000	3.5588	0.3725
3.1884	24.4827	84000	3.5476	0.3731
3.2206	24.7742	85000	3.5405	0.3733
3.1329	25.0656	86000	3.5600	0.3722
3.1725	25.3571	87000	3.5554	0.3726
3.1891	25.6486	88000	3.5475	0.3732
3.2141	25.9401	89000	3.5403	0.3734
3.1573	26.2314	90000	3.5573	0.3726
3.179	26.5229	91000	3.5522	0.3731
3.1943	26.8144	92000	3.5423	0.3736
3.1194	27.1058	93000	3.5631	0.3728
3.1611	27.3973	94000	3.5540	0.3732
3.1853	27.6888	95000	3.5459	0.3736
3.196	27.9803	96000	3.5396	0.3738
3.1423	28.2717	97000	3.5573	0.3730
3.1734	28.5632	98000	3.5503	0.3732
3.1748	28.8547	99000	3.5433	0.3739
3.1103	29.1460	100000	3.5612	0.3727
3.141	29.4375	101000	3.5540	0.3734
3.1626	29.7290	102000	3.5463	0.3738

Framework versions

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