exceptions_exp2_swap_0.3_cost_to_push_2128

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

• Loss: 3.5810
• Accuracy: 0.3661

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: 2128
• 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
4.8475	0.2915	1000	4.7750	0.2502
4.3456	0.5831	2000	4.2943	0.2978
4.1456	0.8746	3000	4.1059	0.3138
3.9944	1.1662	4000	3.9968	0.3234
3.9491	1.4577	5000	3.9227	0.3306
3.8924	1.7493	6000	3.8664	0.3360
3.7617	2.0408	7000	3.8208	0.3401
3.7602	2.3324	8000	3.7899	0.3432
3.7334	2.6239	9000	3.7620	0.3459
3.7346	2.9155	10000	3.7338	0.3483
3.6444	3.2070	11000	3.7238	0.3499
3.659	3.4985	12000	3.7059	0.3517
3.6575	3.7901	13000	3.6863	0.3535
3.5507	4.0816	14000	3.6783	0.3551
3.5711	4.3732	15000	3.6675	0.3562
3.5827	4.6647	16000	3.6550	0.3573
3.5824	4.9563	17000	3.6408	0.3583
3.5035	5.2478	18000	3.6454	0.3591
3.5303	5.5394	19000	3.6323	0.3601
3.5382	5.8309	20000	3.6206	0.3610
3.4441	6.1224	21000	3.6226	0.3609
3.4783	6.4140	22000	3.6176	0.3619
3.5042	6.7055	23000	3.6065	0.3628
3.5019	6.9971	24000	3.5965	0.3634
3.4391	7.2886	25000	3.6066	0.3635
3.4671	7.5802	26000	3.5964	0.3643
3.4646	7.8717	27000	3.5879	0.3648
3.3788	8.1633	28000	3.5949	0.3645
3.42	8.4548	29000	3.5912	0.3657
3.4206	8.7464	30000	3.5810	0.3661
3.3393	9.0379	31000	3.5860	0.3663
3.3852	9.3294	32000	3.5854	0.3662
3.4032	9.6210	33000	3.5760	0.3670
3.4131	9.9125	34000	3.5672	0.3674
3.3514	10.2041	35000	3.5798	0.3672
3.3637	10.4956	36000	3.5698	0.3677
3.3826	10.7872	37000	3.5626	0.3685
3.3083	11.0787	38000	3.5737	0.3679
3.3538	11.3703	39000	3.5722	0.3681
3.363	11.6618	40000	3.5646	0.3687
3.3675	11.9534	41000	3.5520	0.3693
3.3242	12.2449	42000	3.5694	0.3687
3.3425	12.5364	43000	3.5615	0.3693
3.3599	12.8280	44000	3.5545	0.3693
3.2748	13.1195	45000	3.5677	0.3692
3.308	13.4111	46000	3.5632	0.3696
3.3306	13.7026	47000	3.5567	0.3703
3.3501	13.9942	48000	3.5477	0.3704
3.303	14.2857	49000	3.5616	0.3700
3.3112	14.5773	50000	3.5578	0.3704
3.3277	14.8688	51000	3.5486	0.3709
3.2649	15.1603	52000	3.5625	0.3701
3.2859	15.4519	53000	3.5579	0.3706
3.3246	15.7434	54000	3.5483	0.3712
3.2136	16.0350	55000	3.5616	0.3705
3.2559	16.3265	56000	3.5585	0.3707
3.2834	16.6181	57000	3.5516	0.3713
3.2976	16.9096	58000	3.5446	0.3715
3.2242	17.2012	59000	3.5605	0.3707
3.2621	17.4927	60000	3.5490	0.3716
3.2843	17.7843	61000	3.5449	0.3718
3.1983	18.0758	62000	3.5611	0.3714
3.2407	18.3673	63000	3.5544	0.3718
3.2768	18.6589	64000	3.5447	0.3720
3.2854	18.9504	65000	3.5428	0.3721
3.2234	19.2420	66000	3.5554	0.3717
3.2568	19.5335	67000	3.5503	0.3724
3.2694	19.8251	68000	3.5420	0.3728
3.1784	20.1166	69000	3.5584	0.3718
3.2217	20.4082	70000	3.5522	0.3721
3.2423	20.6997	71000	3.5455	0.3725
3.2513	20.9913	72000	3.5366	0.3730
3.2074	21.2828	73000	3.5550	0.3723
3.2366	21.5743	74000	3.5475	0.3726
3.2257	21.8659	75000	3.5367	0.3732
3.1729	22.1574	76000	3.5582	0.3723
3.2132	22.4490	77000	3.5498	0.3726
3.2481	22.7405	78000	3.5411	0.3731
3.1369	23.0321	79000	3.5574	0.3725
3.1759	23.3236	80000	3.5522	0.3729
3.2125	23.6152	81000	3.5454	0.3733
3.2274	23.9067	82000	3.5361	0.3737
3.1613	24.1983	83000	3.5553	0.3729
3.1876	24.4898	84000	3.5515	0.3730
3.2129	24.7813	85000	3.5437	0.3735
3.1389	25.0729	86000	3.5526	0.3732
3.1698	25.3644	87000	3.5525	0.3731
3.1955	25.6560	88000	3.5473	0.3735
3.1939	25.9475	89000	3.5392	0.3737
3.1502	26.2391	90000	3.5545	0.3731
3.1742	26.5306	91000	3.5458	0.3737
3.1846	26.8222	92000	3.5407	0.3741
3.1308	27.1137	93000	3.5584	0.3731
3.1635	27.4052	94000	3.5564	0.3734
3.192	27.6968	95000	3.5428	0.3742
3.186	27.9883	96000	3.5344	0.3744
3.1337	28.2799	97000	3.5569	0.3733
3.1666	28.5714	98000	3.5477	0.3739
3.1893	28.8630	99000	3.5428	0.3742
3.115	29.1545	100000	3.5589	0.3734
3.1486	29.4461	101000	3.5559	0.3735
3.1697	29.7376	102000	3.5468	0.3742
3.0804	30.0292	103000	3.5555	0.3737
3.12	30.3207	104000	3.5551	0.3737
3.1463	30.6122	105000	3.5494	0.3742
3.1608	30.9038	106000	3.5439	0.3744
3.1076	31.1953	107000	3.5588	0.3735
3.1425	31.4869	108000	3.5515	0.3741
3.1472	31.7784	109000	3.5468	0.3745
3.0671	32.0700	110000	3.5595	0.3737
3.1161	32.3615	111000	3.5552	0.3741
3.1468	32.6531	112000	3.5482	0.3746
3.145	32.9446	113000	3.5437	0.3746
3.0995	33.2362	114000	3.5615	0.3737
3.1161	33.5277	115000	3.5526	0.3743
3.1275	33.8192	116000	3.5474	0.3747

Framework versions

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