exceptions_exp2_swap_last_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: 3.5635
• Accuracy: 0.3686

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
4.827	0.2915	1000	4.7442	0.2561
4.3378	0.5830	2000	4.2805	0.2989
4.1451	0.8744	3000	4.1006	0.3143
3.9836	1.1659	4000	3.9921	0.3248
3.9332	1.4573	5000	3.9192	0.3310
3.8862	1.7488	6000	3.8603	0.3358
3.7529	2.0402	7000	3.8196	0.3405
3.7611	2.3317	8000	3.7883	0.3433
3.7368	2.6232	9000	3.7575	0.3464
3.7297	2.9147	10000	3.7324	0.3490
3.6376	3.2061	11000	3.7216	0.3501
3.6394	3.4976	12000	3.7005	0.3522
3.6555	3.7890	13000	3.6821	0.3541
3.5414	4.0804	14000	3.6774	0.3547
3.5802	4.3719	15000	3.6681	0.3563
3.5858	4.6634	16000	3.6515	0.3573
3.5827	4.9549	17000	3.6389	0.3588
3.5181	5.2463	18000	3.6428	0.3589
3.5277	5.5378	19000	3.6310	0.3599
3.5247	5.8293	20000	3.6205	0.3611
3.445	6.1207	21000	3.6242	0.3613
3.4793	6.4121	22000	3.6165	0.3618
3.4918	6.7036	23000	3.6058	0.3626
3.4973	6.9951	24000	3.5968	0.3635
3.4377	7.2865	25000	3.6041	0.3636
3.4504	7.5780	26000	3.5966	0.3643
3.4674	7.8695	27000	3.5892	0.3645
3.3997	8.1609	28000	3.5985	0.3646
3.4222	8.4524	29000	3.5901	0.3649
3.4419	8.7438	30000	3.5819	0.3660
3.3309	9.0353	31000	3.5857	0.3661
3.3823	9.3267	32000	3.5853	0.3662
3.4037	9.6182	33000	3.5777	0.3669
3.422	9.9097	34000	3.5691	0.3670
3.3441	10.2011	35000	3.5808	0.3672
3.3646	10.4926	36000	3.5740	0.3675
3.3899	10.7841	37000	3.5678	0.3678
3.2966	11.0755	38000	3.5803	0.3675
3.3448	11.3670	39000	3.5719	0.3680
3.3518	11.6584	40000	3.5635	0.3686
3.3783	11.9499	41000	3.5601	0.3689
3.3174	12.2413	42000	3.5736	0.3683
3.331	12.5328	43000	3.5644	0.3690
3.3497	12.8243	44000	3.5536	0.3695
3.2706	13.1157	45000	3.5714	0.3693
3.3165	13.4072	46000	3.5613	0.3696
3.3228	13.6987	47000	3.5576	0.3701
3.3354	13.9901	48000	3.5477	0.3704
3.2886	14.2816	49000	3.5643	0.3698
3.3168	14.5730	50000	3.5564	0.3699
3.3205	14.8645	51000	3.5475	0.3709
3.2486	15.1559	52000	3.5627	0.3702
3.2911	15.4474	53000	3.5564	0.3703
3.3163	15.7389	54000	3.5520	0.3709
3.2042	16.0303	55000	3.5626	0.3701
3.2593	16.3218	56000	3.5571	0.3707
3.2926	16.6133	57000	3.5564	0.3709
3.29	16.9047	58000	3.5457	0.3715
3.2094	17.1962	59000	3.5656	0.3707
3.2549	17.4876	60000	3.5537	0.3714
3.2774	17.7791	61000	3.5467	0.3715
3.1996	18.0705	62000	3.5606	0.3710
3.2336	18.3620	63000	3.5559	0.3712
3.2583	18.6535	64000	3.5519	0.3717
3.2779	18.9450	65000	3.5437	0.3720
3.2214	19.2364	66000	3.5598	0.3712
3.2414	19.5279	67000	3.5502	0.3721
3.2705	19.8193	68000	3.5425	0.3725
3.1875	20.1108	69000	3.5633	0.3717
3.22	20.4022	70000	3.5572	0.3717
3.2437	20.6937	71000	3.5455	0.3726
3.2578	20.9852	72000	3.5413	0.3728
3.1906	21.2766	73000	3.5577	0.3719
3.2174	21.5681	74000	3.5517	0.3724
3.2476	21.8596	75000	3.5444	0.3725
3.1745	22.1510	76000	3.5581	0.3721
3.2013	22.4425	77000	3.5578	0.3722
3.23	22.7339	78000	3.5485	0.3726
3.1331	23.0254	79000	3.5606	0.3722
3.1746	23.3168	80000	3.5597	0.3718
3.2066	23.6083	81000	3.5495	0.3726
3.2254	23.8998	82000	3.5419	0.3731
3.1479	24.1912	83000	3.5603	0.3724
3.1789	24.4827	84000	3.5523	0.3728
3.1977	24.7742	85000	3.5453	0.3730
3.1223	25.0656	86000	3.5620	0.3725
3.1664	25.3571	87000	3.5550	0.3726
3.1943	25.6485	88000	3.5484	0.3731
3.197	25.9400	89000	3.5400	0.3738
3.1405	26.2314	90000	3.5603	0.3729
3.188	26.5229	91000	3.5512	0.3730
3.1877	26.8144	92000	3.5454	0.3734
3.1106	27.1058	93000	3.5613	0.3728
3.1566	27.3973	94000	3.5538	0.3731
3.1717	27.6888	95000	3.5519	0.3732
3.1891	27.9802	96000	3.5401	0.3740
3.1227	28.2717	97000	3.5597	0.3732
3.1568	28.5631	98000	3.5465	0.3735
3.1617	28.8546	99000	3.5478	0.3736
3.0999	29.1460	100000	3.5618	0.3729
3.1365	29.4375	101000	3.5549	0.3732
3.1535	29.7290	102000	3.5509	0.3735
3.0911	30.0204	103000	3.5586	0.3734
3.1188	30.3119	104000	3.5617	0.3733
3.1406	30.6034	105000	3.5526	0.3738
3.1614	30.8948	106000	3.5438	0.3742
3.0903	31.1863	107000	3.5605	0.3733
3.1162	31.4777	108000	3.5545	0.3736
3.1424	31.7692	109000	3.5518	0.3738

Framework versions

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