exceptions_exp2_swap_require_to_push_2128

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

• Loss: 3.5539
• Accuracy: 0.3700

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
4.8192	0.2911	1000	0.2564	4.7433
4.3282	0.5822	2000	0.2996	4.2811
4.144	0.8733	3000	0.3153	4.0938
3.9882	1.1642	4000	0.3254	3.9896
3.9317	1.4553	5000	0.3320	3.9145
3.8791	1.7464	6000	0.3372	3.8550
3.7504	2.0373	7000	0.3418	3.8126
3.7635	2.3284	8000	0.3444	3.7833
3.7451	2.6195	9000	0.3469	3.7554
3.7276	2.9106	10000	0.3496	3.7265
3.6278	3.2014	11000	0.3519	3.7129
3.6485	3.4925	12000	0.3534	3.6949
3.6328	3.7837	13000	0.3547	3.6779
3.5314	4.0745	14000	0.3565	3.6718
3.5674	4.3656	15000	0.3574	3.6580
3.5784	4.6567	16000	0.3585	3.6458
3.5636	4.9478	17000	0.3599	3.6340
3.5007	5.2387	18000	0.3602	3.6339
3.5142	5.5298	19000	0.3615	3.6226
3.5268	5.8209	20000	0.3622	3.6115
3.4413	6.1118	21000	0.3629	3.6174
3.4651	6.4029	22000	0.3634	3.6087
3.487	6.6940	23000	0.3641	3.5994
3.4933	6.9851	24000	0.3649	3.5884
3.4254	7.2760	25000	0.3645	3.5965
3.4417	7.5671	26000	0.3657	3.5886
3.4691	7.8582	27000	0.3661	3.5787
3.3747	8.1490	28000	0.3661	3.5894
3.4154	8.4401	29000	0.3667	3.5833
3.426	8.7313	30000	0.3673	3.5738
3.3242	9.0221	31000	0.3673	3.5804
3.3761	9.3132	32000	0.3672	3.5763
3.3917	9.6043	33000	0.3681	3.5689
3.4219	9.8954	34000	0.3687	3.5596
3.3281	10.1863	35000	0.3683	3.5688
3.363	10.4774	36000	0.3689	3.5635
3.3869	10.7685	37000	0.3695	3.5561
3.2757	11.0594	38000	0.3695	3.5678
3.3403	11.3505	39000	0.3695	3.5644
3.3597	11.6416	40000	0.3700	3.5539
3.3754	11.9327	41000	0.3707	3.5451
3.3079	12.2236	42000	0.3701	3.5603
3.3291	12.5147	43000	0.3704	3.5530
3.351	12.8058	44000	0.3711	3.5483
3.2635	13.0966	45000	0.3705	3.5575
3.3017	13.3878	46000	0.3708	3.5550
3.331	13.6789	47000	0.3713	3.5486
3.3337	13.9700	48000	0.3715	3.5398
3.2713	14.2608	49000	0.3709	3.5591
3.3054	14.5519	50000	0.3717	3.5483
3.3194	14.8430	51000	0.3722	3.5419
3.2275	15.1339	52000	0.3715	3.5551
3.2847	15.4250	53000	0.3717	3.5485
3.2896	15.7161	54000	0.3723	3.5415
3.2594	16.0070	55000	0.3720	3.5495
3.2474	16.2981	56000	0.3720	3.5505
3.2715	16.5892	57000	0.3729	3.5415
3.3027	16.8803	58000	0.3729	3.5350
3.2205	17.1712	59000	0.3724	3.5492
3.2666	17.4623	60000	0.3725	3.5432
3.2936	17.7534	61000	0.3731	3.5360
3.1937	18.0442	62000	0.3726	3.5481
3.2302	18.3354	63000	0.3729	3.5461
3.2573	18.6265	64000	0.3731	3.5399
3.2744	18.9176	65000	0.3740	3.5306
3.2095	19.2084	66000	0.3728	3.5492
3.2518	19.4995	67000	0.3734	3.5391
3.2435	19.7906	68000	0.3737	3.5335
3.1722	20.0815	69000	0.3732	3.5475
3.2126	20.3726	70000	0.3736	3.5439
3.2362	20.6637	71000	0.3741	3.5342
3.2486	20.9548	72000	0.3745	3.5283
3.1904	21.2457	73000	0.3736	3.5439
3.2111	21.5368	74000	0.3741	3.5391
3.2275	21.8279	75000	0.3743	3.5325
3.1541	22.1188	76000	0.3737	3.5487
3.1968	22.4099	77000	0.3739	3.5436
3.2183	22.7010	78000	0.3743	3.5365
3.2421	22.9921	79000	0.3749	3.5262
3.1805	23.2830	80000	0.3739	3.5468
3.1668	23.5741	81000	3.5471	0.3739
3.1977	23.8652	82000	3.5403	0.3744
3.1545	24.1563	83000	3.5513	0.3740
3.1834	24.4474	84000	3.5444	0.3741
3.2117	24.7385	85000	3.5345	0.3747
3.1165	25.0294	86000	3.5460	0.3742
3.1622	25.3205	87000	3.5466	0.3742
3.1757	25.6116	88000	3.5385	0.3745
3.2034	25.9027	89000	3.5300	0.3752
3.1307	26.1936	90000	3.5470	0.3740
3.1562	26.4847	91000	3.5443	0.3747
3.1927	26.7758	92000	3.5336	0.3753
3.1075	27.0667	93000	3.5485	0.3743
3.1492	27.3578	94000	3.5436	0.3749
3.1663	27.6489	95000	3.5358	0.3749
3.1839	27.9400	96000	3.5292	0.3757
3.1223	28.2308	97000	3.5485	0.3746
3.1447	28.5219	98000	3.5409	0.3751
3.1618	28.8131	99000	3.5339	0.3755
3.1018	29.1039	100000	3.5488	0.3743

Framework versions

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