exceptions_exp2_swap_last_to_drop_2128

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

• Loss: 3.5591
• 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: 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	Validation Loss	Accuracy
4.8526	0.2915	1000	4.7764	0.2513
4.3512	0.5830	2000	4.2893	0.2989
4.1542	0.8744	3000	4.1047	0.3145
3.9999	1.1659	4000	3.9948	0.3242
3.9347	1.4573	5000	3.9202	0.3308
3.8879	1.7488	6000	3.8624	0.3361
3.7429	2.0402	7000	3.8215	0.3405
3.7571	2.3317	8000	3.7907	0.3436
3.7466	2.6232	9000	3.7587	0.3463
3.7288	2.9147	10000	3.7345	0.3487
3.627	3.2061	11000	3.7215	0.3505
3.6524	3.4976	12000	3.7017	0.3521
3.6369	3.7890	13000	3.6860	0.3539
3.5531	4.0804	14000	3.6763	0.3554
3.5603	4.3719	15000	3.6623	0.3566
3.5848	4.6634	16000	3.6517	0.3574
3.5819	4.9549	17000	3.6365	0.3589
3.5155	5.2463	18000	3.6393	0.3593
3.521	5.5378	19000	3.6291	0.3602
3.5333	5.8293	20000	3.6180	0.3612
3.4502	6.1207	21000	3.6245	0.3614
3.469	6.4121	22000	3.6160	0.3620
3.4894	6.7036	23000	3.6061	0.3629
3.5104	6.9951	24000	3.5958	0.3637
3.4433	7.2865	25000	3.6044	0.3636
3.4634	7.5780	26000	3.5937	0.3645
3.4602	7.8695	27000	3.5860	0.3649
3.4127	8.1609	28000	3.5980	0.3647
3.4196	8.4524	29000	3.5890	0.3652
3.4306	8.7438	30000	3.5791	0.3660
3.3403	9.0353	31000	3.5886	0.3661
3.3899	9.3267	32000	3.5825	0.3662
3.3959	9.6182	33000	3.5750	0.3667
3.4164	9.9097	34000	3.5676	0.3675
3.338	10.2011	35000	3.5817	0.3672
3.3855	10.4926	36000	3.5759	0.3676
3.3901	10.7841	37000	3.5650	0.3683
3.2958	11.0755	38000	3.5741	0.3681
3.3405	11.3670	39000	3.5687	0.3685
3.3551	11.6584	40000	3.5591	0.3690
3.3772	11.9499	41000	3.5534	0.3693
3.2932	12.2413	42000	3.5702	0.3689
3.3407	12.5328	43000	3.5643	0.3690
3.3408	12.8243	44000	3.5559	0.3696
3.2729	13.1157	45000	3.5659	0.3696
3.3098	13.4072	46000	3.5617	0.3694
3.3322	13.6987	47000	3.5574	0.3703
3.3399	13.9901	48000	3.5479	0.3705
3.2819	14.2816	49000	3.5625	0.3701
3.313	14.5730	50000	3.5558	0.3703
3.3264	14.8645	51000	3.5499	0.3709
3.2558	15.1559	52000	3.5630	0.3703
3.2871	15.4474	53000	3.5571	0.3707
3.3118	15.7389	54000	3.5499	0.3709
3.2039	16.0303	55000	3.5621	0.3706
3.2593	16.3218	56000	3.5584	0.3706
3.2808	16.6133	57000	3.5515	0.3714
3.2929	16.9047	58000	3.5430	0.3719
3.2243	17.1962	59000	3.5616	0.3714
3.2663	17.4876	60000	3.5550	0.3714
3.2768	17.7791	61000	3.5454	0.3721
3.2049	18.0705	62000	3.5570	0.3711
3.2479	18.3620	63000	3.5553	0.3716
3.2741	18.6535	64000	3.5461	0.3720
3.2736	18.9450	65000	3.5410	0.3722
3.2073	19.2364	66000	3.5582	0.3719
3.2386	19.5279	67000	3.5517	0.3721
3.2592	19.8193	68000	3.5428	0.3726
3.1774	20.1108	69000	3.5593	0.3720
3.2074	20.4022	70000	3.5550	0.3718
3.2424	20.6937	71000	3.5471	0.3724
3.2744	20.9852	72000	3.5414	0.3728
3.2009	21.2766	73000	3.5555	0.3722
3.2148	21.5681	74000	3.5505	0.3726
3.2211	21.8596	75000	3.5429	0.3729
3.1644	22.1510	76000	3.5601	0.3724
3.2155	22.4425	77000	3.5517	0.3726
3.225	22.7339	78000	3.5439	0.3731
3.1301	23.0254	79000	3.5566	0.3725
3.179	23.3168	80000	3.5565	0.3725
3.2085	23.6083	81000	3.5468	0.3730
3.2333	23.8998	82000	3.5390	0.3734
3.1682	24.1912	83000	3.5626	0.3727
3.1713	24.4827	84000	3.5500	0.3731
3.214	24.7742	85000	3.5406	0.3735
3.1206	25.0656	86000	3.5589	0.3728
3.1728	25.3571	87000	3.5511	0.3730
3.1922	25.6485	88000	3.5463	0.3736
3.226	25.9400	89000	3.5421	0.3738
3.1439	26.2314	90000	3.5622	0.3727
3.1825	26.5229	91000	3.5491	0.3735
3.175	26.8144	92000	3.5438	0.3736
3.1202	27.1058	93000	3.5602	0.3730
3.1623	27.3973	94000	3.5533	0.3734
3.1896	27.6888	95000	3.5450	0.3739
3.1956	27.9802	96000	3.5413	0.3742
3.1249	28.2717	97000	3.5601	0.3731
3.1643	28.5631	98000	3.5527	0.3735
3.1807	28.8546	99000	3.5438	0.3739
3.1035	29.1460	100000	3.5584	0.3732
3.1371	29.4375	101000	3.5553	0.3735
3.1526	29.7290	102000	3.5468	0.3741

Framework versions

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