exceptions_exp2_last_to_drop_frequency_1032

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

• Loss: 3.5565
• Accuracy: 0.3699

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	Validation Loss	Accuracy
4.8164	0.2912	1000	4.7452	0.2562
4.339	0.5824	2000	4.2831	0.3002
4.1293	0.8736	3000	4.0976	0.3151
3.9912	1.1645	4000	3.9954	0.3248
3.9401	1.4557	5000	3.9215	0.3314
3.8958	1.7469	6000	3.8627	0.3368
3.7519	2.0379	7000	3.8213	0.3405
3.756	2.3290	8000	3.7892	0.3437
3.7507	2.6202	9000	3.7573	0.3468
3.731	2.9114	10000	3.7307	0.3490
3.6279	3.2024	11000	3.7205	0.3507
3.6457	3.4936	12000	3.7006	0.3528
3.6444	3.7848	13000	3.6810	0.3545
3.5475	4.0757	14000	3.6758	0.3558
3.5642	4.3669	15000	3.6626	0.3569
3.5793	4.6581	16000	3.6514	0.3577
3.5751	4.9493	17000	3.6344	0.3596
3.5111	5.2402	18000	3.6383	0.3600
3.5112	5.5314	19000	3.6262	0.3604
3.5203	5.8226	20000	3.6176	0.3618
3.4341	6.1136	21000	3.6186	0.3627
3.478	6.4048	22000	3.6096	0.3629
3.4906	6.6959	23000	3.6027	0.3635
3.4966	6.9871	24000	3.5950	0.3647
3.4272	7.2781	25000	3.5991	0.3646
3.4526	7.5693	26000	3.5938	0.3650
3.4506	7.8605	27000	3.5811	0.3662
3.3616	8.1514	28000	3.5915	0.3656
3.4033	8.4426	29000	3.5843	0.3666
3.4251	8.7338	30000	3.5736	0.3672
3.3199	9.0248	31000	3.5795	0.3674
3.3801	9.3159	32000	3.5797	0.3671
3.3891	9.6071	33000	3.5694	0.3680
3.4101	9.8983	34000	3.5642	0.3685
3.3363	10.1893	35000	3.5748	0.3681
3.3778	10.4805	36000	3.5685	0.3686
3.3797	10.7716	37000	3.5581	0.3693
3.2924	11.0626	38000	3.5690	0.3691
3.3359	11.3538	39000	3.5671	0.3693
3.3503	11.6450	40000	3.5565	0.3699
3.3774	11.9362	41000	3.5490	0.3706
3.3027	12.2271	42000	3.5610	0.3696
3.3386	12.5183	43000	3.5549	0.3702
3.3472	12.8095	44000	3.5485	0.3710
3.2543	13.1005	45000	3.5629	0.3704
3.2975	13.3916	46000	3.5622	0.3705
3.3194	13.6828	47000	3.5483	0.3712
3.3425	13.9740	48000	3.5404	0.3717
3.2744	14.2650	49000	3.5580	0.3713
3.3138	14.5562	50000	3.5478	0.3717
3.32	14.8474	51000	3.5402	0.3720
3.2496	15.1383	52000	3.5571	0.3714
3.2803	15.4295	53000	3.5499	0.3717
3.2919	15.7207	54000	3.5405	0.3721
3.2175	16.0116	55000	3.5516	0.3721
3.2547	16.3028	56000	3.5515	0.3722
3.2692	16.5940	57000	3.5416	0.3726
3.2916	16.8852	58000	3.5351	0.3730
3.2282	17.1762	59000	3.5533	0.3721
3.2584	17.4674	60000	3.5436	0.3727
3.2724	17.7585	61000	3.5372	0.3732
3.1868	18.0495	62000	3.5490	0.3726
3.2329	18.3407	63000	3.5462	0.3726
3.2426	18.6319	64000	3.5440	0.3727
3.2577	18.9231	65000	3.5335	0.3736
3.1949	19.2140	66000	3.5489	0.3729
3.2359	19.5052	67000	3.5423	0.3731
3.2688	19.7964	68000	3.5396	0.3735
3.1636	20.0874	69000	3.5508	0.3729
3.2153	20.3785	70000	3.5451	0.3733
3.2345	20.6697	71000	3.5408	0.3734
3.2576	20.9609	72000	3.5285	0.3742
3.1837	21.2519	73000	3.5471	0.3736
3.2214	21.5431	74000	3.5406	0.3739
3.2441	21.8343	75000	3.5327	0.3744
3.162	22.1252	76000	3.5475	0.3737
3.1943	22.4164	77000	3.5435	0.3737
3.2206	22.7076	78000	3.5353	0.3743
3.2293	22.9988	79000	3.5264	0.3749
3.1645	23.2897	80000	3.5482	0.3736
3.201	23.5809	81000	3.5388	0.3742
3.2211	23.8721	82000	3.5299	0.3747
3.1515	24.1631	83000	3.5483	0.3742
3.1878	24.4543	84000	3.5453	0.3740
3.2018	24.7454	85000	3.5341	0.3745
3.1168	25.0364	86000	3.5502	0.3742
3.158	25.3276	87000	3.5451	0.3743
3.1849	25.6188	88000	3.5404	0.3747
3.1933	25.9100	89000	3.5328	0.3751
3.1394	26.2009	90000	3.5515	0.3742
3.1698	26.4921	91000	3.5400	0.3749
3.1911	26.7833	92000	3.5383	0.3749
3.099	27.0743	93000	3.5501	0.3746
3.1493	27.3654	94000	3.5436	0.3746
3.1648	27.6566	95000	3.5364	0.3751
3.1793	27.9478	96000	3.5336	0.3751
3.1201	28.2388	97000	3.5506	0.3747
3.1451	28.5300	98000	3.5380	0.3752
3.1765	28.8212	99000	3.5373	0.3750

Framework versions

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