exceptions_exp2_swap_0.7_cost_to_push_40817

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

• Loss: 3.5674
• Accuracy: 0.3684

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
4.8317	0.2917	1000	4.7685	0.2521
4.3567	0.5834	2000	4.2898	0.2977
4.1606	0.8750	3000	4.1113	0.3140
4.0129	1.1665	4000	3.9987	0.3240
3.9405	1.4582	5000	3.9221	0.3309
3.8841	1.7499	6000	3.8658	0.3359
3.7634	2.0414	7000	3.8191	0.3401
3.7623	2.3331	8000	3.7927	0.3433
3.7502	2.6248	9000	3.7612	0.3459
3.7213	2.9165	10000	3.7356	0.3486
3.6448	3.2080	11000	3.7245	0.3500
3.647	3.4996	12000	3.7068	0.3520
3.6607	3.7913	13000	3.6865	0.3537
3.5566	4.0828	14000	3.6816	0.3546
3.5605	4.3745	15000	3.6713	0.3559
3.5792	4.6662	16000	3.6548	0.3572
3.5696	4.9579	17000	3.6423	0.3584
3.5114	5.2494	18000	3.6454	0.3586
3.5205	5.5411	19000	3.6344	0.3596
3.5466	5.8327	20000	3.6221	0.3605
3.4599	6.1243	21000	3.6298	0.3610
3.4778	6.4159	22000	3.6188	0.3618
3.4998	6.7076	23000	3.6094	0.3622
3.5083	6.9993	24000	3.5975	0.3632
3.4246	7.2908	25000	3.6066	0.3633
3.4786	7.5825	26000	3.6000	0.3640
3.4675	7.8742	27000	3.5897	0.3647
3.3767	8.1657	28000	3.5985	0.3644
3.4312	8.4574	29000	3.5922	0.3650
3.442	8.7490	30000	3.5836	0.3658
3.3449	9.0405	31000	3.5919	0.3653
3.3924	9.3322	32000	3.5859	0.3661
3.4126	9.6239	33000	3.5795	0.3666
3.4162	9.9156	34000	3.5690	0.3673
3.3491	10.2071	35000	3.5839	0.3665
3.3755	10.4988	36000	3.5739	0.3672
3.397	10.7905	37000	3.5678	0.3680
3.3174	11.0820	38000	3.5801	0.3672
3.3318	11.3736	39000	3.5742	0.3674
3.3597	11.6653	40000	3.5674	0.3684
3.3895	11.9570	41000	3.5598	0.3689
3.334	12.2485	42000	3.5707	0.3684
3.3507	12.5402	43000	3.5640	0.3690
3.3587	12.8319	44000	3.5582	0.3692
3.2946	13.1234	45000	3.5682	0.3690
3.3207	13.4151	46000	3.5656	0.3692
3.3373	13.7067	47000	3.5574	0.3697
3.3516	13.9984	48000	3.5520	0.3703
3.301	14.2899	49000	3.5653	0.3693
3.3261	14.5816	50000	3.5611	0.3699
3.3368	14.8733	51000	3.5524	0.3705
3.2546	15.1648	52000	3.5666	0.3697
3.2967	15.4565	53000	3.5616	0.3702
3.3185	15.7482	54000	3.5524	0.3704
3.2039	16.0397	55000	3.5613	0.3704
3.2676	16.3313	56000	3.5596	0.3705
3.2911	16.6230	57000	3.5556	0.3706
3.3108	16.9147	58000	3.5478	0.3711
3.2358	17.2062	59000	3.5619	0.3707
3.2703	17.4979	60000	3.5540	0.3712
3.2906	17.7896	61000	3.5512	0.3713
3.1964	18.0811	62000	3.5649	0.3708
3.2472	18.3728	63000	3.5604	0.3710
3.2495	18.6644	64000	3.5500	0.3717
3.2898	18.9561	65000	3.5430	0.3720
3.2198	19.2476	66000	3.5600	0.3713
3.2476	19.5393	67000	3.5538	0.3715
3.2666	19.8310	68000	3.5440	0.3721
3.1929	20.1225	69000	3.5635	0.3714
3.2104	20.4142	70000	3.5590	0.3713
3.2562	20.7059	71000	3.5455	0.3720
3.2735	20.9975	72000	3.5410	0.3727
3.2084	21.2891	73000	3.5614	0.3716
3.2305	21.5807	74000	3.5505	0.3722
3.2621	21.8724	75000	3.5427	0.3725
3.1855	22.1639	76000	3.5604	0.3720
3.218	22.4556	77000	3.5500	0.3723
3.2207	22.7473	78000	3.5457	0.3730
3.1523	23.0388	79000	3.5590	0.3721
3.2082	23.3305	80000	3.5572	0.3722
3.2118	23.6222	81000	3.5476	0.3727
3.219	23.9138	82000	3.5416	0.3732
3.1723	24.2053	83000	3.5599	0.3722
3.1917	24.4970	84000	3.5534	0.3730
3.2127	24.7887	85000	3.5454	0.3731
3.1432	25.0802	86000	3.5624	0.3723
3.1864	25.3719	87000	3.5579	0.3726
3.2011	25.6636	88000	3.5502	0.3730
3.2208	25.9553	89000	3.5391	0.3736
3.1548	26.2468	90000	3.5585	0.3727
3.1786	26.5384	91000	3.5479	0.3731
3.1971	26.8301	92000	3.5435	0.3734
3.1445	27.1216	93000	3.5622	0.3724
3.1676	27.4133	94000	3.5564	0.3731
3.1797	27.7050	95000	3.5484	0.3734
3.2029	27.9967	96000	3.5410	0.3735
3.1359	28.2882	97000	3.5566	0.3731
3.1758	28.5799	98000	3.5510	0.3733
3.1902	28.8715	99000	3.5432	0.3737
3.1161	29.1630	100000	3.5599	0.3728
3.1439	29.4547	101000	3.5518	0.3734
3.1657	29.7464	102000	3.5451	0.3738
3.0772	30.0379	103000	3.5603	0.3732
3.1417	30.3296	104000	3.5578	0.3732
3.1599	30.6213	105000	3.5503	0.3738
3.1691	30.9130	106000	3.5438	0.3740
3.0977	31.2045	107000	3.5578	0.3731
3.1382	31.4961	108000	3.5545	0.3737
3.1528	31.7878	109000	3.5468	0.3739

Framework versions

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