exceptions_exp2_swap_0.3_last_to_hit_1032

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

• Loss: 3.5644
• Accuracy: 0.3687

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
4.8328	0.2915	1000	4.7665	0.2527
4.3617	0.5830	2000	4.2908	0.2982
4.1524	0.8745	3000	4.1078	0.3140
4.0083	1.1659	4000	4.0004	0.3234
3.9504	1.4574	5000	3.9245	0.3305
3.8894	1.7488	6000	3.8678	0.3359
3.7521	2.0402	7000	3.8210	0.3400
3.7519	2.3317	8000	3.7926	0.3433
3.7443	2.6232	9000	3.7610	0.3460
3.7312	2.9147	10000	3.7368	0.3484
3.6456	3.2061	11000	3.7237	0.3506
3.6476	3.4976	12000	3.7047	0.3522
3.6584	3.7891	13000	3.6853	0.3536
3.5575	4.0805	14000	3.6799	0.3550
3.5672	4.3719	15000	3.6660	0.3563
3.5873	4.6634	16000	3.6547	0.3571
3.5795	4.9549	17000	3.6403	0.3584
3.5132	5.2463	18000	3.6417	0.3593
3.5129	5.5378	19000	3.6309	0.3602
3.5437	5.8293	20000	3.6218	0.3609
3.4471	6.1207	21000	3.6235	0.3616
3.485	6.4122	22000	3.6188	0.3618
3.4862	6.7037	23000	3.6070	0.3631
3.5077	6.9952	24000	3.5992	0.3634
3.4277	7.2865	25000	3.6059	0.3632
3.4628	7.5780	26000	3.5976	0.3642
3.4685	7.8695	27000	3.5884	0.3648
3.3723	8.1609	28000	3.5968	0.3648
3.4126	8.4524	29000	3.5903	0.3655
3.4439	8.7439	30000	3.5815	0.3659
3.3293	9.0353	31000	3.5867	0.3661
3.3883	9.3268	32000	3.5858	0.3661
3.391	9.6183	33000	3.5755	0.3667
3.4114	9.9098	34000	3.5697	0.3673
3.3473	10.2011	35000	3.5827	0.3669
3.3641	10.4926	36000	3.5755	0.3676
3.3843	10.7841	37000	3.5657	0.3683
3.3037	11.0755	38000	3.5790	0.3679
3.3446	11.3670	39000	3.5693	0.3682
3.3659	11.6585	40000	3.5644	0.3687
3.3688	11.9500	41000	3.5572	0.3692
3.3236	12.2414	42000	3.5705	0.3686
3.3415	12.5329	43000	3.5655	0.3692
3.3573	12.8243	44000	3.5571	0.3696
3.2796	13.1157	45000	3.5706	0.3690
3.3065	13.4072	46000	3.5622	0.3698
3.3301	13.6987	47000	3.5528	0.3700
3.3421	13.9902	48000	3.5473	0.3705
3.2825	14.2816	49000	3.5631	0.3696
3.3223	14.5731	50000	3.5567	0.3703
3.3347	14.8646	51000	3.5476	0.3708
3.2572	15.1559	52000	3.5646	0.3702
3.2928	15.4474	53000	3.5589	0.3704
3.3015	15.7389	54000	3.5496	0.3708
3.22	16.0303	55000	3.5636	0.3707
3.2637	16.3218	56000	3.5567	0.3709
3.2838	16.6133	57000	3.5526	0.3713
3.3057	16.9048	58000	3.5414	0.3717
3.227	17.1962	59000	3.5621	0.3711
3.2597	17.4877	60000	3.5515	0.3712
3.2879	17.7792	61000	3.5458	0.3720
3.202	18.0705	62000	3.5570	0.3714
3.2395	18.3620	63000	3.5552	0.3715
3.2611	18.6535	64000	3.5494	0.3720
3.2813	18.9450	65000	3.5410	0.3725
3.2294	19.2364	66000	3.5566	0.3714
3.2541	19.5279	67000	3.5482	0.3721
3.2583	19.8194	68000	3.5429	0.3725
3.1785	20.1108	69000	3.5596	0.3717
3.2258	20.4023	70000	3.5554	0.3721
3.2423	20.6938	71000	3.5435	0.3726
3.2558	20.9853	72000	3.5396	0.3729
3.2062	21.2766	73000	3.5547	0.3723
3.223	21.5681	74000	3.5481	0.3726
3.2448	21.8596	75000	3.5427	0.3731
3.1669	22.1510	76000	3.5594	0.3721
3.2084	22.4425	77000	3.5505	0.3727
3.2245	22.7340	78000	3.5465	0.3730
3.1337	23.0254	79000	3.5574	0.3727
3.1966	23.3169	80000	3.5544	0.3726
3.2172	23.6083	81000	3.5474	0.3730
3.2341	23.8998	82000	3.5399	0.3735
3.158	24.1912	83000	3.5568	0.3729
3.1912	24.4827	84000	3.5507	0.3728
3.2156	24.7742	85000	3.5407	0.3735
3.1401	25.0656	86000	3.5583	0.3729
3.1712	25.3571	87000	3.5571	0.3730
3.1944	25.6486	88000	3.5510	0.3732
3.2055	25.9401	89000	3.5363	0.3740
3.1546	26.2314	90000	3.5572	0.3732
3.1739	26.5229	91000	3.5511	0.3731
3.2081	26.8144	92000	3.5455	0.3736
3.1203	27.1058	93000	3.5569	0.3731
3.1599	27.3973	94000	3.5533	0.3733
3.1767	27.6888	95000	3.5460	0.3739
3.1897	27.9803	96000	3.5376	0.3743
3.1287	28.2717	97000	3.5572	0.3731
3.173	28.5632	98000	3.5503	0.3736
3.1799	28.8547	99000	3.5419	0.3744
3.116	29.1460	100000	3.5599	0.3735
3.1425	29.4375	101000	3.5556	0.3737
3.1707	29.7290	102000	3.5477	0.3741
3.0922	30.0204	103000	3.5590	0.3736
3.1302	30.3119	104000	3.5595	0.3735
3.1527	30.6034	105000	3.5491	0.3741
3.1692	30.8949	106000	3.5419	0.3744
3.1204	31.1863	107000	3.5605	0.3736
3.1335	31.4778	108000	3.5534	0.3741
3.1511	31.7693	109000	3.5460	0.3744

Framework versions

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