exceptions_exp2_swap_require_to_hit_40817

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: 40817
• 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
4.8268	0.2911	1000	4.7465	0.2558
4.3357	0.5822	2000	4.2782	0.3000
4.1426	0.8733	3000	4.0941	0.3157
4.001	1.1642	4000	3.9901	0.3252
3.9346	1.4553	5000	3.9106	0.3326
3.8828	1.7464	6000	3.8530	0.3376
3.7536	2.0373	7000	3.8106	0.3417
3.7423	2.3284	8000	3.7821	0.3446
3.7415	2.6195	9000	3.7527	0.3472
3.7266	2.9106	10000	3.7274	0.3498
3.6342	3.2014	11000	3.7148	0.3514
3.6458	3.4925	12000	3.6962	0.3534
3.6462	3.7837	13000	3.6776	0.3547
3.5396	4.0745	14000	3.6715	0.3562
3.5516	4.3656	15000	3.6618	0.3572
3.5805	4.6567	16000	3.6468	0.3587
3.5772	4.9478	17000	3.6335	0.3598
3.4968	5.2387	18000	3.6380	0.3599
3.5073	5.5298	19000	3.6263	0.3610
3.5311	5.8209	20000	3.6140	0.3618
3.4358	6.1118	21000	3.6165	0.3622
3.4734	6.4029	22000	3.6106	0.3630
3.4889	6.6940	23000	3.6004	0.3638
3.4885	6.9851	24000	3.5916	0.3647
3.4359	7.2760	25000	3.5977	0.3647
3.4439	7.5671	26000	3.5917	0.3651
3.4508	7.8582	27000	3.5805	0.3661
3.3949	8.1490	28000	3.5917	0.3656
3.3978	8.4401	29000	3.5822	0.3664
3.4259	8.7313	30000	3.5762	0.3671
3.3152	9.0221	31000	3.5793	0.3673
3.3879	9.3132	32000	3.5794	0.3673
3.3938	9.6043	33000	3.5695	0.3679
3.4115	9.8954	34000	3.5598	0.3685
3.3308	10.1863	35000	3.5733	0.3681
3.3627	10.4774	36000	3.5669	0.3686
3.3765	10.7685	37000	3.5624	0.3691
3.2937	11.0594	38000	3.5701	0.3690
3.3279	11.3505	39000	3.5671	0.3695
3.3719	11.6416	40000	3.5560	0.3697
3.3639	11.9327	41000	3.5487	0.3707
3.302	12.2236	42000	3.5643	0.3698
3.3366	12.5147	43000	3.5555	0.3705
3.3445	12.8058	44000	3.5515	0.3708
3.2828	13.0966	45000	3.5621	0.3702
3.2953	13.3878	46000	3.5556	0.3706
3.3078	13.6789	47000	3.5520	0.3711
3.3446	13.9700	48000	3.5401	0.3715
3.2724	14.2608	49000	3.5578	0.3706
3.2974	14.5519	50000	3.5515	0.3715
3.3178	14.8430	51000	3.5447	0.3718
3.2375	15.1339	52000	3.5569	0.3712
3.2772	15.4250	53000	3.5535	0.3715
3.2992	15.7161	54000	3.5445	0.3723
3.2511	16.0070	55000	3.5538	0.3717
3.2394	16.2981	56000	3.5526	0.3718
3.2747	16.5892	57000	3.5445	0.3723
3.2794	16.8803	58000	3.5395	0.3731
3.2266	17.1712	59000	3.5543	0.3721
3.2536	17.4623	60000	3.5503	0.3722
3.2686	17.7534	61000	3.5364	0.3730
3.182	18.0442	62000	3.5529	0.3724
3.2291	18.3354	63000	3.5503	0.3724
3.2581	18.6265	64000	3.5413	0.3729
3.2736	18.9176	65000	3.5354	0.3736
3.2005	19.2084	66000	3.5549	0.3726
3.2281	19.4995	67000	3.5459	0.3729
3.2467	19.7906	68000	3.5360	0.3736
3.1684	20.0815	69000	3.5533	0.3727
3.2171	20.3726	70000	3.5492	0.3729
3.2207	20.6637	71000	3.5399	0.3736
3.2511	20.9548	72000	3.5340	0.3741
3.1786	21.2457	73000	3.5510	0.3735
3.2068	21.5368	74000	3.5429	0.3736
3.2242	21.8279	75000	3.5347	0.3741
3.1702	22.1188	76000	3.5513	0.3734
3.1939	22.4099	77000	3.5459	0.3734
3.2103	22.7010	78000	3.5385	0.3737
3.2235	22.9921	79000	3.5328	0.3743
3.1857	23.2830	80000	3.5497	0.3735
3.2037	23.5741	81000	3.5415	0.3743
3.2284	23.8652	82000	3.5330	0.3744
3.1397	24.1560	83000	3.5506	0.3737
3.1771	24.4471	84000	3.5453	0.3743
3.2014	24.7382	85000	3.5367	0.3746
3.1061	25.0291	86000	3.5539	0.3739
3.1642	25.3202	87000	3.5483	0.3739
3.1839	25.6113	88000	3.5437	0.3745
3.2017	25.9024	89000	3.5319	0.3748
3.1412	26.1933	90000	3.5569	0.3737
3.1635	26.4844	91000	3.5422	0.3744
3.1822	26.7755	92000	3.5379	0.3751
3.1221	27.0664	93000	3.5531	0.3743
3.1453	27.3575	94000	3.5506	0.3742
3.1801	27.6486	95000	3.5437	0.3749
3.1735	27.9397	96000	3.5351	0.3751
3.1263	28.2306	97000	3.5537	0.3742
3.1597	28.5217	98000	3.5454	0.3746
3.17	28.8128	99000	3.5355	0.3754
3.0868	29.1036	100000	3.5491	0.3745
3.131	29.3947	101000	3.5500	0.3743
3.1589	29.6858	102000	3.5411	0.3751
3.1768	29.9769	103000	3.5321	0.3755
3.1074	30.2678	104000	3.5514	0.3745
3.1342	30.5589	105000	3.5426	0.3754
3.1485	30.8500	106000	3.5370	0.3755
3.0909	31.1409	107000	3.5540	0.3746
3.1148	31.4320	108000	3.5466	0.3750
3.1434	31.7231	109000	3.5404	0.3754

Framework versions

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