exceptions_exp2_swap_last_to_carry_5039

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

• Loss: 3.5643
• Accuracy: 0.3686

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
4.8281	0.2915	1000	4.7547	0.2546
4.345	0.5830	2000	4.2857	0.2982
4.1443	0.8744	3000	4.1021	0.3145
4.001	1.1659	4000	3.9960	0.3243
3.9522	1.4573	5000	3.9178	0.3311
3.8807	1.7488	6000	3.8626	0.3361
3.7594	2.0402	7000	3.8202	0.3401
3.7556	2.3317	8000	3.7893	0.3435
3.7484	2.6232	9000	3.7585	0.3459
3.727	2.9147	10000	3.7313	0.3486
3.6405	3.2061	11000	3.7218	0.3502
3.6512	3.4976	12000	3.7028	0.3526
3.6587	3.7890	13000	3.6833	0.3538
3.5487	4.0804	14000	3.6784	0.3555
3.5628	4.3719	15000	3.6658	0.3562
3.5778	4.6634	16000	3.6522	0.3574
3.5771	4.9549	17000	3.6389	0.3585
3.5106	5.2463	18000	3.6398	0.3593
3.5282	5.5378	19000	3.6315	0.3602
3.5289	5.8293	20000	3.6178	0.3610
3.4447	6.1207	21000	3.6233	0.3611
3.4696	6.4121	22000	3.6152	0.3622
3.491	6.7036	23000	3.6062	0.3629
3.5021	6.9951	24000	3.5967	0.3637
3.428	7.2865	25000	3.6032	0.3637
3.4482	7.5780	26000	3.5957	0.3645
3.4753	7.8695	27000	3.5880	0.3650
3.391	8.1609	28000	3.5953	0.3647
3.4028	8.4524	29000	3.5874	0.3655
3.4299	8.7438	30000	3.5801	0.3661
3.3336	9.0353	31000	3.5858	0.3660
3.3799	9.3267	32000	3.5827	0.3666
3.4013	9.6182	33000	3.5726	0.3672
3.416	9.9097	34000	3.5681	0.3676
3.3389	10.2011	35000	3.5779	0.3673
3.3688	10.4926	36000	3.5745	0.3674
3.3823	10.7841	37000	3.5648	0.3686
3.2921	11.0755	38000	3.5733	0.3681
3.3445	11.3670	39000	3.5688	0.3683
3.3643	11.6584	40000	3.5643	0.3686
3.3842	11.9499	41000	3.5540	0.3694
3.3067	12.2413	42000	3.5680	0.3691
3.3341	12.5328	43000	3.5616	0.3695
3.3593	12.8243	44000	3.5548	0.3698
3.27	13.1157	45000	3.5671	0.3694
3.3132	13.4072	46000	3.5620	0.3697
3.3266	13.6987	47000	3.5526	0.3703
3.3381	13.9901	48000	3.5490	0.3705
3.29	14.2816	49000	3.5590	0.3701
3.3029	14.5730	50000	3.5557	0.3703
3.319	14.8645	51000	3.5467	0.3712
3.2517	15.1559	52000	3.5620	0.3702
3.2919	15.4474	53000	3.5562	0.3708
3.3008	15.7389	54000	3.5473	0.3713
3.2056	16.0303	55000	3.5601	0.3708
3.2621	16.3218	56000	3.5581	0.3709
3.2888	16.6133	57000	3.5519	0.3715
3.2936	16.9047	58000	3.5442	0.3716
3.2229	17.1962	59000	3.5602	0.3712
3.2601	17.4876	60000	3.5543	0.3715
3.2825	17.7791	61000	3.5448	0.3722
3.203	18.0705	62000	3.5562	0.3716
3.248	18.3620	63000	3.5520	0.3717
3.2554	18.6535	64000	3.5459	0.3722
3.2808	18.9450	65000	3.5401	0.3724
3.2191	19.2364	66000	3.5577	0.3718
3.2393	19.5279	67000	3.5503	0.3721
3.2559	19.8193	68000	3.5438	0.3729
3.2003	20.1108	69000	3.5574	0.3717
3.2114	20.4022	70000	3.5477	0.3723
3.2391	20.6937	71000	3.5449	0.3726
3.2672	20.9852	72000	3.5370	0.3729
3.2086	21.2766	73000	3.5550	0.3724
3.2184	21.5681	74000	3.5478	0.3727
3.2388	21.8596	75000	3.5428	0.3732
3.1662	22.1510	76000	3.5581	0.3726
3.1885	22.4425	77000	3.5509	0.3727
3.233	22.7339	78000	3.5418	0.3731
3.1175	23.0254	79000	3.5574	0.3725
3.1748	23.3168	80000	3.5539	0.3727
3.204	23.6083	81000	3.5426	0.3735
3.213	23.8998	82000	3.5390	0.3737
3.153	24.1912	83000	3.5549	0.3728
3.1885	24.4827	84000	3.5486	0.3732
3.2148	24.7742	85000	3.5441	0.3737
3.1338	25.0656	86000	3.5564	0.3727
3.165	25.3571	87000	3.5535	0.3732
3.2024	25.6485	88000	3.5462	0.3734
3.2005	25.9400	89000	3.5367	0.3741
3.1555	26.2314	90000	3.5563	0.3730
3.1641	26.5229	91000	3.5486	0.3735
3.1968	26.8144	92000	3.5411	0.3740
3.1267	27.1058	93000	3.5581	0.3733
3.1492	27.3973	94000	3.5505	0.3737
3.1847	27.6888	95000	3.5452	0.3739
3.1852	27.9802	96000	3.5366	0.3745
3.1411	28.2717	97000	3.5580	0.3732
3.1613	28.5631	98000	3.5468	0.3743
3.1805	28.8546	99000	3.5411	0.3743
3.1074	29.1460	100000	3.5580	0.3735
3.1537	29.4375	101000	3.5520	0.3736
3.1579	29.7290	102000	3.5465	0.3744
3.0861	30.0204	103000	3.5541	0.3735
3.1277	30.3119	104000	3.5551	0.3740
3.1496	30.6034	105000	3.5498	0.3740
3.1623	30.8948	106000	3.5410	0.3744
3.0949	31.1863	107000	3.5601	0.3737
3.1309	31.4777	108000	3.5537	0.3740
3.1487	31.7692	109000	3.5441	0.3745
3.0706	32.0606	110000	3.5579	0.3739
3.1172	32.3521	111000	3.5538	0.3741
3.1377	32.6436	112000	3.5499	0.3745
3.141	32.9351	113000	3.5398	0.3750
3.085	33.2265	114000	3.5570	0.3741
3.1239	33.5180	115000	3.5519	0.3745
3.1422	33.8094	116000	3.5458	0.3744

Framework versions

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