exceptions_exp2_swap_last_to_hit_5039

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

• Loss: 3.5632
• Accuracy: 0.3688

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: 5039
• 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
4.8207	0.2915	1000	4.7540	0.2549
4.3442	0.5830	2000	4.2829	0.2991
4.1444	0.8744	3000	4.1005	0.3150
4.0008	1.1659	4000	3.9931	0.3248
3.951	1.4573	5000	3.9175	0.3314
3.8806	1.7488	6000	3.8604	0.3365
3.7583	2.0402	7000	3.8184	0.3404
3.7557	2.3317	8000	3.7857	0.3436
3.7466	2.6232	9000	3.7564	0.3460
3.7265	2.9147	10000	3.7295	0.3489
3.6393	3.2061	11000	3.7186	0.3506
3.6507	3.4976	12000	3.7009	0.3526
3.6577	3.7890	13000	3.6832	0.3538
3.5491	4.0804	14000	3.6753	0.3555
3.5617	4.3719	15000	3.6647	0.3562
3.5783	4.6634	16000	3.6515	0.3574
3.5766	4.9549	17000	3.6386	0.3585
3.5106	5.2463	18000	3.6394	0.3593
3.5274	5.5378	19000	3.6324	0.3601
3.5281	5.8293	20000	3.6192	0.3608
3.4447	6.1207	21000	3.6229	0.3613
3.4698	6.4121	22000	3.6146	0.3622
3.491	6.7036	23000	3.6052	0.3630
3.5026	6.9951	24000	3.5948	0.3636
3.4297	7.2865	25000	3.6031	0.3639
3.4495	7.5780	26000	3.5955	0.3644
3.476	7.8695	27000	3.5868	0.3651
3.3922	8.1609	28000	3.5942	0.3650
3.4042	8.4524	29000	3.5877	0.3654
3.4309	8.7438	30000	3.5799	0.3662
3.335	9.0353	31000	3.5834	0.3662
3.3806	9.3267	32000	3.5829	0.3666
3.4014	9.6182	33000	3.5740	0.3671
3.4179	9.9097	34000	3.5683	0.3675
3.3393	10.2011	35000	3.5780	0.3672
3.3697	10.4926	36000	3.5725	0.3673
3.3846	10.7841	37000	3.5659	0.3683
3.2923	11.0755	38000	3.5723	0.3680
3.3456	11.3670	39000	3.5719	0.3683
3.3652	11.6584	40000	3.5632	0.3688
3.3842	11.9499	41000	3.5536	0.3695
3.3079	12.2413	42000	3.5690	0.3691
3.3347	12.5328	43000	3.5624	0.3694
3.3593	12.8243	44000	3.5547	0.3699
3.27	13.1157	45000	3.5660	0.3693
3.314	13.4072	46000	3.5610	0.3696
3.3283	13.6987	47000	3.5550	0.3700
3.3389	13.9901	48000	3.5469	0.3706
3.2891	14.2816	49000	3.5622	0.3699
3.3043	14.5730	50000	3.5536	0.3707
3.3204	14.8645	51000	3.5468	0.3710
3.2525	15.1559	52000	3.5632	0.3702
3.2929	15.4474	53000	3.5569	0.3707
3.3019	15.7389	54000	3.5507	0.3711
3.2065	16.0303	55000	3.5618	0.3707
3.2647	16.3218	56000	3.5573	0.3710
3.2899	16.6133	57000	3.5509	0.3713
3.2942	16.9047	58000	3.5450	0.3715
3.2247	17.1962	59000	3.5602	0.3710
3.2604	17.4876	60000	3.5542	0.3714
3.2848	17.7791	61000	3.5435	0.3720
3.2024	18.0705	62000	3.5599	0.3713
3.2495	18.3620	63000	3.5516	0.3718
3.2572	18.6535	64000	3.5491	0.3719
3.2816	18.9450	65000	3.5392	0.3724
3.2196	19.2364	66000	3.5593	0.3715
3.2423	19.5279	67000	3.5504	0.3722
3.2577	19.8193	68000	3.5434	0.3726
3.2012	20.1108	69000	3.5570	0.3719
3.2126	20.4022	70000	3.5497	0.3722
3.2409	20.6937	71000	3.5453	0.3725
3.2687	20.9852	72000	3.5377	0.3729
3.21	21.2766	73000	3.5562	0.3722
3.2196	21.5681	74000	3.5497	0.3726
3.2404	21.8596	75000	3.5402	0.3732
3.1685	22.1510	76000	3.5557	0.3722
3.1897	22.4425	77000	3.5523	0.3727
3.2342	22.7339	78000	3.5429	0.3731
3.1185	23.0254	79000	3.5570	0.3728
3.1749	23.3168	80000	3.5556	0.3725
3.2044	23.6083	81000	3.5455	0.3730
3.2149	23.8998	82000	3.5364	0.3736
3.1541	24.1912	83000	3.5546	0.3726
3.1897	24.4827	84000	3.5477	0.3731
3.2161	24.7742	85000	3.5438	0.3736
3.1347	25.0656	86000	3.5551	0.3727
3.1665	25.3571	87000	3.5540	0.3731
3.2038	25.6485	88000	3.5477	0.3734
3.2029	25.9400	89000	3.5378	0.3739
3.1574	26.2314	90000	3.5559	0.3730
3.1658	26.5229	91000	3.5493	0.3733
3.1994	26.8144	92000	3.5430	0.3737
3.128	27.1058	93000	3.5585	0.3731
3.1511	27.3973	94000	3.5525	0.3735
3.1874	27.6888	95000	3.5444	0.3740
3.1862	27.9802	96000	3.5386	0.3742
3.1422	28.2717	97000	3.5561	0.3733
3.1638	28.5631	98000	3.5488	0.3740
3.1814	28.8546	99000	3.5433	0.3742
3.1099	29.1460	100000	3.5587	0.3733
3.1549	29.4375	101000	3.5529	0.3735
3.1596	29.7290	102000	3.5491	0.3740

Framework versions

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