exceptions_exp2_swap_require_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: 3.5560
• Accuracy: 0.3697

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	Accuracy	Validation Loss
4.8313	0.2911	1000	0.2540	4.7551
4.327	0.5822	2000	0.2997	4.2808
4.143	0.8733	3000	0.3158	4.0936
3.9875	1.1642	4000	0.3258	3.9878
3.9281	1.4553	5000	0.3324	3.9116
3.8772	1.7464	6000	0.3377	3.8521
3.7465	2.0373	7000	0.3420	3.8094
3.7597	2.3284	8000	0.3449	3.7793
3.7427	2.6195	9000	0.3474	3.7533
3.7251	2.9106	10000	0.3499	3.7231
3.629	3.2017	11000	0.3521	3.7163
3.6484	3.4928	12000	0.3533	3.6963
3.6332	3.7839	13000	0.3548	3.6787
3.5315	4.0748	14000	0.3562	3.6728
3.5681	4.3659	15000	0.3573	3.6608
3.5781	4.6570	16000	0.3587	3.6463
3.5613	4.9481	17000	0.3599	3.6338
3.4987	5.2390	18000	0.3604	3.6328
3.5154	5.5301	19000	0.3614	3.6225
3.528	5.8212	20000	0.3622	3.6133
3.442	6.1121	21000	0.3626	3.6160
3.4671	6.4032	22000	0.3636	3.6099
3.4893	6.6943	23000	0.3643	3.5987
3.4985	6.9854	24000	0.3649	3.5882
3.4264	7.2763	25000	0.3644	3.5983
3.4448	7.5674	26000	0.3653	3.5921
3.4691	7.8585	27000	0.3662	3.5819
3.3775	8.1493	28000	0.3663	3.5903
3.4149	8.4404	29000	0.3668	3.5849
3.4252	8.7315	30000	0.3674	3.5762
3.3288	9.0224	31000	0.3674	3.5799
3.3776	9.3135	32000	0.3673	3.5804
3.3921	9.6046	33000	0.3678	3.5719
3.4212	9.8957	34000	0.3686	3.5636
3.3329	10.1866	35000	0.3683	3.5705
3.367	10.4777	36000	0.3687	3.5656
3.3889	10.7688	37000	0.3696	3.5565
3.2762	11.0597	38000	0.3693	3.5687
3.3424	11.3508	39000	0.3693	3.5665
3.3651	11.6419	40000	0.3697	3.5560
3.3767	11.9330	41000	0.3705	3.5478
3.3124	12.2239	42000	0.3699	3.5633
3.3323	12.5150	43000	0.3704	3.5532
3.3498	12.8061	44000	0.3707	3.5505
3.2658	13.0969	45000	0.3704	3.5606
3.3054	13.3880	46000	0.3709	3.5550
3.3295	13.6791	47000	0.3710	3.5496
3.3345	13.9702	48000	0.3717	3.5407
3.2725	14.2611	49000	0.3708	3.5595
3.3044	14.5522	50000	0.3714	3.5515
3.3212	14.8433	51000	0.3720	3.5415
3.2308	15.1342	52000	0.3712	3.5574
3.288	15.4253	53000	0.3717	3.5501
3.2921	15.7164	54000	0.3723	3.5417
3.2591	16.0073	55000	0.3717	3.5503
3.2485	16.2984	56000	0.3718	3.5517
3.2739	16.5895	57000	0.3727	3.5459
3.3028	16.8806	58000	0.3731	3.5335
3.2237	17.1715	59000	0.3720	3.5547
3.2656	17.4626	60000	0.3724	3.5480
3.2953	17.7537	61000	0.3731	3.5424
3.1951	18.0445	62000	0.3726	3.5493
3.2331	18.3356	63000	0.3729	3.5486
3.2593	18.6267	64000	0.3727	3.5409
3.2761	18.9179	65000	0.3735	3.5339
3.2115	19.2087	66000	0.3726	3.5528
3.253	19.4998	67000	0.3732	3.5418
3.2466	19.7909	68000	0.3734	3.5372
3.1752	20.0818	69000	0.3730	3.5488
3.2125	20.3729	70000	0.3733	3.5475
3.2385	20.6640	71000	0.3739	3.5358
3.251	20.9551	72000	0.3739	3.5311
3.1939	21.2460	73000	0.3729	3.5486
3.2151	21.5371	74000	0.3736	3.5441
3.2295	21.8282	75000	0.3741	3.5372
3.1566	22.1191	76000	0.3733	3.5541
3.2004	22.4102	77000	0.3737	3.5459
3.2214	22.7013	78000	0.3741	3.5404
3.2437	22.9924	79000	0.3744	3.5299
3.1845	23.2832	80000	0.3739	3.5493
3.1994	23.5743	81000	0.3744	3.5441
3.2281	23.8655	82000	0.3746	3.5324
3.1537	24.1563	83000	0.3739	3.5519
3.1843	24.4474	84000	0.3741	3.5443
3.2124	24.7385	85000	0.3745	3.5374
3.1189	25.0294	86000	0.3743	3.5461
3.1644	25.3205	87000	0.3739	3.5504
3.1798	25.6116	88000	0.3746	3.5400
3.2071	25.9027	89000	0.3752	3.5329
3.1338	26.1936	90000	0.3739	3.5499
3.1636	26.4844	91000	3.5494	0.3741
3.1776	26.7755	92000	3.5446	0.3746
3.1153	27.0667	93000	3.5517	0.3739
3.1558	27.3578	94000	3.5487	0.3742
3.1684	27.6489	95000	3.5416	0.3746
3.1868	27.9400	96000	3.5338	0.3754
3.1243	28.2308	97000	3.5556	0.3741
3.149	28.5219	98000	3.5435	0.3750
3.1667	28.8131	99000	3.5367	0.3752
3.1054	29.1039	100000	3.5510	0.3743
3.1242	29.3950	101000	3.5459	0.3747
3.168	29.6861	102000	3.5423	0.3751
3.1654	29.9772	103000	3.5371	0.3755
3.1044	30.2681	104000	3.5516	0.3749
3.135	30.5592	105000	3.5479	0.3747
3.1584	30.8503	106000	3.5359	0.3756
3.0948	31.1412	107000	3.5543	0.3746
3.1236	31.4323	108000	3.5485	0.3751
3.1344	31.7234	109000	3.5397	0.3755
3.0595	32.0143	110000	3.5503	0.3749

Framework versions

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