exceptions_exp2_swap_0.7_resemble_to_drop_1032

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

• Loss: 3.5662
• Accuracy: 0.3683

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: 1032
• 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
4.8421	0.2915	1000	0.2539	4.7622
4.356	0.5831	2000	0.2985	4.2935
4.1548	0.8746	3000	0.3143	4.1058
4.017	1.1662	4000	0.3236	3.9993
3.9376	1.4577	5000	0.3307	3.9279
3.8874	1.7493	6000	0.3357	3.8649
3.7712	2.0408	7000	0.3400	3.8225
3.7637	2.3324	8000	0.3428	3.7959
3.7525	2.6239	9000	0.3456	3.7631
3.7308	2.9155	10000	0.3483	3.7371
3.6435	3.2070	11000	0.3500	3.7252
3.6634	3.4985	12000	0.3518	3.7072
3.6609	3.7901	13000	0.3532	3.6884
3.5535	4.0816	14000	0.3547	3.6797
3.5846	4.3732	15000	0.3557	3.6698
3.5857	4.6647	16000	0.3568	3.6575
3.5668	4.9563	17000	0.3581	3.6438
3.5188	5.2478	18000	0.3588	3.6447
3.5443	5.5394	19000	0.3598	3.6332
3.5363	5.8309	20000	0.3607	3.6232
3.4558	6.1224	21000	0.3610	3.6260
3.4914	6.4140	22000	0.3616	3.6189
3.5046	6.7055	23000	0.3626	3.6095
3.5148	6.9971	24000	0.3631	3.6001
3.4412	7.2886	25000	0.3632	3.6070
3.4779	7.5802	26000	0.3639	3.6000
3.4714	7.8717	27000	0.3644	3.5894
3.3867	8.1633	28000	0.3645	3.6019
3.4306	8.4548	29000	0.3651	3.5922
3.4539	8.7464	30000	0.3657	3.5809
3.3412	9.0379	31000	0.3657	3.5896
3.3828	9.3294	32000	0.3659	3.5875
3.4198	9.6210	33000	0.3669	3.5790
3.4339	9.9125	34000	0.3670	3.5723
3.3578	10.2041	35000	0.3667	3.5832
3.384	10.4956	36000	0.3672	3.5763
3.396	10.7872	37000	0.3677	3.5681
3.3103	11.0787	38000	0.3675	3.5809
3.35	11.3703	39000	0.3675	3.5738
3.3797	11.6618	40000	0.3683	3.5662
3.3819	11.9534	41000	0.3687	3.5588
3.3023	12.2449	42000	0.3685	3.5722
3.3478	12.5364	43000	0.3687	3.5637
3.3609	12.8280	44000	0.3694	3.5553
3.2814	13.1195	45000	0.3688	3.5694
3.3283	13.4111	46000	0.3691	3.5633
3.3333	13.7026	47000	0.3699	3.5573
3.3471	13.9942	48000	0.3704	3.5456
3.2933	14.2857	49000	0.3697	3.5612
3.3184	14.5773	50000	0.3700	3.5579
3.3417	14.8688	51000	0.3704	3.5482
3.2583	15.1603	52000	0.3699	3.5657
3.2864	15.4519	53000	0.3704	3.5597
3.3149	15.7434	54000	0.3707	3.5500
3.2158	16.0350	55000	0.3705	3.5616
3.2654	16.3265	56000	0.3705	3.5567
3.2883	16.6181	57000	0.3708	3.5539
3.3078	16.9096	58000	0.3713	3.5461
3.2323	17.2012	59000	0.3705	3.5604
3.2821	17.4927	60000	0.3711	3.5531
3.2906	17.7843	61000	0.3712	3.5486
3.2125	18.0758	62000	0.3707	3.5622
3.2485	18.3673	63000	0.3714	3.5551
3.2758	18.6589	64000	0.3716	3.5509
3.277	18.9504	65000	0.3722	3.5407
3.2186	19.2420	66000	0.3712	3.5590
3.2452	19.5335	67000	0.3718	3.5501
3.2819	19.8251	68000	0.3722	3.5411
3.1866	20.1166	69000	0.3716	3.5631
3.2367	20.4082	70000	0.3719	3.5527
3.256	20.6997	71000	0.3720	3.5458
3.2627	20.9913	72000	0.3725	3.5410
3.2108	21.2828	73000	0.3719	3.5577
3.2428	21.5743	74000	0.3721	3.5496
3.2478	21.8659	75000	0.3727	3.5416
3.1858	22.1574	76000	0.3720	3.5563
3.2168	22.4490	77000	0.3724	3.5525
3.2357	22.7405	78000	0.3727	3.5436
3.1474	23.0321	79000	0.3723	3.5575
3.1963	23.3236	80000	0.3725	3.5532
3.1907	23.6152	81000	3.5608	0.3718
3.2165	23.9067	82000	3.5509	0.3727
3.1731	24.1983	83000	3.5604	0.3719
3.2112	24.4898	84000	3.5523	0.3727
3.2331	24.7813	85000	3.5460	0.3731
3.1384	25.0729	86000	3.5545	0.3726
3.1709	25.3644	87000	3.5560	0.3725
3.2031	25.6560	88000	3.5457	0.3728
3.2162	25.9475	89000	3.5394	0.3735
3.1572	26.2391	90000	3.5569	0.3727
3.1872	26.5306	91000	3.5523	0.3729
3.2053	26.8222	92000	3.5431	0.3736
3.1282	27.1137	93000	3.5576	0.3729
3.1673	27.4052	94000	3.5515	0.3731
3.1803	27.6968	95000	3.5458	0.3735
3.2022	27.9883	96000	3.5421	0.3739
3.1435	28.2799	97000	3.5602	0.3728
3.165	28.5714	98000	3.5526	0.3734
3.178	28.8630	99000	3.5424	0.3739
3.1168	29.1545	100000	3.5616	0.3730
3.168	29.4461	101000	3.5545	0.3735
3.1809	29.7376	102000	3.5470	0.3736
3.087	30.0292	103000	3.5612	0.3731
3.1311	30.3207	104000	3.5576	0.3733
3.1605	30.6122	105000	3.5522	0.3737
3.1753	30.9038	106000	3.5429	0.3744
3.1019	31.1953	107000	3.5630	0.3730
3.1443	31.4869	108000	3.5535	0.3738
3.1674	31.7784	109000	3.5450	0.3742

Framework versions

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