exceptions_exp2_swap_require_to_drop_5039

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

• Loss: 3.5547
• Accuracy: 0.3701

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: 5039
• 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
4.8279	0.2911	1000	0.2558	4.7454
4.3342	0.5822	2000	0.2993	4.2822
4.1427	0.8733	3000	0.3158	4.0932
3.9879	1.1642	4000	0.3250	3.9889
3.9259	1.4553	5000	0.3321	3.9127
3.8683	1.7464	6000	0.3370	3.8549
3.7463	2.0373	7000	0.3415	3.8130
3.748	2.3284	8000	0.3445	3.7832
3.7327	2.6195	9000	0.3474	3.7517
3.723	2.9106	10000	0.3497	3.7273
3.6333	3.2014	11000	0.3518	3.7129
3.6459	3.4925	12000	0.3537	3.6937
3.6396	3.7837	13000	0.3549	3.6775
3.5385	4.0745	14000	0.3563	3.6677
3.5632	4.3656	15000	0.3575	3.6571
3.5758	4.6567	16000	0.3589	3.6429
3.581	4.9478	17000	0.3601	3.6306
3.5044	5.2387	18000	0.3607	3.6313
3.5205	5.5298	19000	0.3615	3.6234
3.5243	5.8209	20000	0.3625	3.6114
3.4455	6.1118	21000	0.3625	3.6156
3.4616	6.4029	22000	0.3633	3.6086
3.4781	6.6940	23000	0.3642	3.5985
3.4903	6.9851	24000	0.3647	3.5888
3.4108	7.2760	25000	0.3649	3.5966
3.4453	7.5671	26000	0.3655	3.5889
3.4592	7.8582	27000	0.3665	3.5783
3.3725	8.1490	28000	0.3664	3.5865
3.4095	8.4401	29000	0.3668	3.5800
3.4187	8.7313	30000	0.3673	3.5703
3.3167	9.0221	31000	0.3671	3.5771
3.3618	9.3132	32000	0.3679	3.5781
3.3949	9.6043	33000	0.3682	3.5686
3.4173	9.8954	34000	0.3688	3.5582
3.3165	10.1863	35000	0.3683	3.5730
3.3582	10.4774	36000	0.3688	3.5658
3.377	10.7685	37000	0.3693	3.5576
3.2762	11.0594	38000	0.3692	3.5663
3.3352	11.3505	39000	0.3695	3.5628
3.357	11.6416	40000	0.3701	3.5547
3.3651	11.9327	41000	0.3708	3.5478
3.2958	12.2236	42000	0.3699	3.5645
3.325	12.5147	43000	0.3703	3.5578
3.3449	12.8058	44000	0.3710	3.5468
3.2516	13.0966	45000	0.3706	3.5610
3.2936	13.3878	46000	0.3708	3.5574
3.3268	13.6789	47000	0.3714	3.5490
3.3467	13.9700	48000	0.3719	3.5401
3.2773	14.2608	49000	0.3711	3.5561
3.3127	14.5519	50000	0.3717	3.5468
3.3241	14.8430	51000	0.3720	3.5421
3.2303	15.1339	52000	0.3719	3.5544
3.2817	15.4250	53000	0.3722	3.5463
3.2936	15.7161	54000	0.3722	3.5436
3.2458	16.0070	55000	0.3720	3.5525
3.256	16.2981	56000	0.3722	3.5498
3.2708	16.5892	57000	0.3727	3.5419
3.283	16.8803	58000	0.3733	3.5343
3.2243	17.1712	59000	0.3725	3.5513
3.2465	17.4623	60000	0.3728	3.5447
3.2745	17.7534	61000	0.3732	3.5361
3.1858	18.0442	62000	0.3728	3.5493
3.2391	18.3354	63000	0.3723	3.5484
3.2426	18.6265	64000	0.3734	3.5398
3.2714	18.9176	65000	0.3738	3.5322
3.2148	19.2084	66000	0.3732	3.5500
3.2418	19.4995	67000	0.3735	3.5424
3.2534	19.7906	68000	0.3742	3.5333
3.1632	20.0815	69000	0.3732	3.5508
3.2168	20.3726	70000	0.3737	3.5436
3.2446	20.6637	71000	0.3740	3.5385
3.2515	20.9548	72000	0.3744	3.5304
3.1864	21.2457	73000	0.3734	3.5474
3.2199	21.5368	74000	0.3742	3.5408
3.2271	21.8279	75000	0.3743	3.5319
3.1653	22.1188	76000	0.3738	3.5493
3.1894	22.4099	77000	0.3739	3.5443
3.2196	22.7010	78000	0.3746	3.5371
3.2487	22.9921	79000	0.3747	3.5306
3.1854	23.2830	80000	0.3741	3.5466
3.1849	23.5741	81000	3.5461	0.3741
3.1958	23.8652	82000	3.5426	0.3741
3.1534	24.1563	83000	3.5528	0.3737
3.184	24.4474	84000	3.5439	0.3744
3.2089	24.7385	85000	3.5364	0.3747
3.1137	25.0294	86000	3.5495	0.3743
3.1475	25.3205	87000	3.5475	0.3740
3.1837	25.6116	88000	3.5390	0.3749
3.1943	25.9027	89000	3.5335	0.3753
3.1439	26.1936	90000	3.5503	0.3741
3.1672	26.4847	91000	3.5413	0.3748
3.1849	26.7758	92000	3.5381	0.3749
3.0921	27.0667	93000	3.5524	0.3744
3.1447	27.3578	94000	3.5462	0.3748
3.1592	27.6489	95000	3.5415	0.3749
3.1866	27.9400	96000	3.5333	0.3755
3.1215	28.2308	97000	3.5529	0.3749
3.1396	28.5219	98000	3.5423	0.3752
3.1494	28.8131	99000	3.5356	0.3754
3.0925	29.1039	100000	3.5490	0.3748

Framework versions

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