exceptions_exp2_last_to_carry_frequency_1032

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

• Loss: 3.5603
• Accuracy: 0.3693

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
4.8127	0.2913	1000	4.7312	0.2579
4.3304	0.5826	2000	4.2763	0.3000
4.1529	0.8739	3000	4.0966	0.3153
3.9983	1.1652	4000	3.9904	0.3250
3.9249	1.4565	5000	3.9173	0.3314
3.8903	1.7477	6000	3.8589	0.3369
3.7635	2.0390	7000	3.8180	0.3408
3.7702	2.3303	8000	3.7853	0.3442
3.7354	2.6216	9000	3.7552	0.3471
3.7273	2.9129	10000	3.7297	0.3495
3.6258	3.2042	11000	3.7159	0.3515
3.6523	3.4955	12000	3.6990	0.3531
3.6454	3.7868	13000	3.6826	0.3547
3.5384	4.0781	14000	3.6727	0.3560
3.5642	4.3694	15000	3.6624	0.3569
3.5822	4.6606	16000	3.6478	0.3582
3.5878	4.9519	17000	3.6379	0.3592
3.5126	5.2432	18000	3.6379	0.3600
3.5281	5.5345	19000	3.6286	0.3609
3.5384	5.8258	20000	3.6171	0.3618
3.4365	6.1171	21000	3.6180	0.3622
3.477	6.4084	22000	3.6142	0.3626
3.4841	6.6997	23000	3.6056	0.3636
3.493	6.9910	24000	3.5955	0.3642
3.4211	7.2823	25000	3.6025	0.3645
3.4501	7.5736	26000	3.5919	0.3650
3.4646	7.8648	27000	3.5827	0.3657
3.382	8.1561	28000	3.5925	0.3657
3.4023	8.4474	29000	3.5831	0.3661
3.4206	8.7387	30000	3.5779	0.3670
3.3284	9.0300	31000	3.5839	0.3669
3.3706	9.3213	32000	3.5810	0.3672
3.4008	9.6126	33000	3.5719	0.3678
3.4016	9.9039	34000	3.5663	0.3684
3.344	10.1952	35000	3.5757	0.3680
3.3658	10.4865	36000	3.5698	0.3684
3.3864	10.7777	37000	3.5615	0.3688
3.292	11.0690	38000	3.5681	0.3690
3.3273	11.3603	39000	3.5691	0.3689
3.3604	11.6516	40000	3.5603	0.3693
3.3662	11.9429	41000	3.5509	0.3703
3.3145	12.2342	42000	3.5626	0.3693
3.3408	12.5255	43000	3.5566	0.3703
3.3589	12.8168	44000	3.5519	0.3702
3.2725	13.1081	45000	3.5610	0.3702
3.318	13.3994	46000	3.5582	0.3703
3.342	13.6906	47000	3.5490	0.3712
3.3467	13.9819	48000	3.5447	0.3715
3.2671	14.2732	49000	3.5590	0.3705
3.3074	14.5645	50000	3.5491	0.3712
3.3175	14.8558	51000	3.5418	0.3718
3.2397	15.1471	52000	3.5566	0.3714
3.2702	15.4384	53000	3.5540	0.3715
3.2823	15.7297	54000	3.5401	0.3721
3.2059	16.0210	55000	3.5513	0.3719
3.2616	16.3123	56000	3.5520	0.3720
3.2886	16.6036	57000	3.5430	0.3725
3.3	16.8948	58000	3.5367	0.3728
3.2212	17.1861	59000	3.5522	0.3720
3.2577	17.4774	60000	3.5459	0.3724
3.274	17.7687	61000	3.5362	0.3730
3.191	18.0600	62000	3.5555	0.3721
3.2522	18.3513	63000	3.5486	0.3727
3.2584	18.6426	64000	3.5407	0.3729
3.276	18.9339	65000	3.5341	0.3733
3.2164	19.2252	66000	3.5519	0.3724
3.2382	19.5165	67000	3.5460	0.3732
3.2541	19.8077	68000	3.5347	0.3737
3.1753	20.0990	69000	3.5532	0.3732
3.2181	20.3903	70000	3.5479	0.3732
3.2277	20.6816	71000	3.5395	0.3736
3.254	20.9729	72000	3.5303	0.3741
3.1843	21.2642	73000	3.5451	0.3734
3.2214	21.5555	74000	3.5389	0.3737
3.2321	21.8468	75000	3.5346	0.3741
3.1704	22.1381	76000	3.5458	0.3735
3.1986	22.4294	77000	3.5432	0.3738
3.2118	22.7207	78000	3.5374	0.3741
3.145	23.0119	79000	3.5437	0.3740
3.1741	23.3032	80000	3.5464	0.3739
3.1999	23.5945	81000	3.5405	0.3744
3.2158	23.8858	82000	3.5319	0.3746
3.1511	24.1771	83000	3.5495	0.3742
3.1905	24.4684	84000	3.5426	0.3743
3.1898	24.7597	85000	3.5386	0.3747
3.1296	25.0510	86000	3.5496	0.3739
3.1601	25.3423	87000	3.5479	0.3742
3.1941	25.6336	88000	3.5339	0.3748
3.2105	25.9248	89000	3.5294	0.3751
3.1418	26.2161	90000	3.5476	0.3742
3.1615	26.5074	91000	3.5427	0.3745
3.1996	26.7987	92000	3.5348	0.3749
3.1119	27.0900	93000	3.5491	0.3745
3.1546	27.3813	94000	3.5435	0.3746
3.1648	27.6726	95000	3.5378	0.3748
3.1949	27.9639	96000	3.5324	0.3755
3.1301	28.2552	97000	3.5506	0.3745
3.1508	28.5465	98000	3.5417	0.3750
3.1696	28.8378	99000	3.5343	0.3751
3.1182	29.1290	100000	3.5521	0.3746
3.1367	29.4203	101000	3.5465	0.3749
3.1549	29.7116	102000	3.5365	0.3752
3.1447	30.0029	103000	3.5441	0.3751
3.1155	30.2942	104000	3.5475	0.3748
3.136	30.5855	105000	3.5417	0.3753
3.1496	30.8768	106000	3.5334	0.3756
3.0895	31.1681	107000	3.5510	0.3749
3.1302	31.4594	108000	3.5446	0.3753
3.1378	31.7507	109000	3.5379	0.3755

Framework versions

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