exceptions_exp2_cost_to_hit_frequency_1032

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

• Loss: 3.5595
• Accuracy: 0.3699

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
4.8199	0.2913	1000	4.7286	0.2583
4.3303	0.5826	2000	4.2863	0.2996
4.1458	0.8739	3000	4.0950	0.3157
3.9913	1.1652	4000	3.9922	0.3252
3.9403	1.4565	5000	3.9162	0.3316
3.8738	1.7478	6000	3.8581	0.3372
3.7607	2.0390	7000	3.8170	0.3415
3.7538	2.3303	8000	3.7856	0.3444
3.7389	2.6216	9000	3.7552	0.3469
3.7227	2.9130	10000	3.7284	0.3496
3.6345	3.2042	11000	3.7158	0.3513
3.651	3.4955	12000	3.6981	0.3531
3.6375	3.7868	13000	3.6834	0.3543
3.5478	4.0781	14000	3.6710	0.3561
3.5756	4.3694	15000	3.6604	0.3571
3.5855	4.6607	16000	3.6468	0.3586
3.5811	4.9520	17000	3.6346	0.3596
3.5085	5.2432	18000	3.6379	0.3602
3.5343	5.5345	19000	3.6267	0.3608
3.5223	5.8259	20000	3.6158	0.3618
3.4387	6.1171	21000	3.6203	0.3623
3.4844	6.4084	22000	3.6126	0.3628
3.4801	6.6997	23000	3.6029	0.3639
3.4928	6.9910	24000	3.5933	0.3648
3.434	7.2823	25000	3.6025	0.3643
3.4633	7.5736	26000	3.5915	0.3652
3.4652	7.8649	27000	3.5824	0.3659
3.3938	8.1561	28000	3.5901	0.3657
3.4171	8.4474	29000	3.5859	0.3659
3.4328	8.7388	30000	3.5773	0.3669
3.3374	9.0300	31000	3.5828	0.3671
3.3887	9.3213	32000	3.5805	0.3673
3.4039	9.6126	33000	3.5734	0.3677
3.4186	9.9039	34000	3.5653	0.3685
3.3404	10.1952	35000	3.5763	0.3682
3.3729	10.4865	36000	3.5691	0.3682
3.3936	10.7778	37000	3.5624	0.3686
3.2871	11.0690	38000	3.5698	0.3692
3.3399	11.3603	39000	3.5649	0.3693
3.3684	11.6517	40000	3.5595	0.3699
3.3729	11.9430	41000	3.5530	0.3702
3.3031	12.2342	42000	3.5656	0.3692
3.3536	12.5255	43000	3.5577	0.3701
3.3609	12.8168	44000	3.5498	0.3705
3.2677	13.1081	45000	3.5631	0.3702
3.3209	13.3994	46000	3.5595	0.3704
3.3413	13.6907	47000	3.5529	0.3709
3.348	13.9820	48000	3.5454	0.3715
3.2852	14.2732	49000	3.5590	0.3707
3.3133	14.5646	50000	3.5517	0.3713
3.3249	14.8559	51000	3.5443	0.3717
3.2419	15.1471	52000	3.5618	0.3710
3.2753	15.4384	53000	3.5520	0.3713
3.3033	15.7297	54000	3.5458	0.3720
3.2057	16.0210	55000	3.5570	0.3718
3.2579	16.3123	56000	3.5514	0.3718
3.282	16.6036	57000	3.5490	0.3721
3.3136	16.8949	58000	3.5383	0.3727
3.2328	17.1861	59000	3.5542	0.3720
3.2686	17.4775	60000	3.5456	0.3724
3.2832	17.7688	61000	3.5407	0.3726
3.1855	18.0600	62000	3.5547	0.3723
3.2353	18.3513	63000	3.5536	0.3721
3.265	18.6426	64000	3.5431	0.3731
3.2902	18.9339	65000	3.5362	0.3731
3.2075	19.2252	66000	3.5558	0.3725
3.2463	19.5165	67000	3.5461	0.3726
3.2592	19.8078	68000	3.5384	0.3731
3.1898	20.0990	69000	3.5559	0.3727
3.2148	20.3904	70000	3.5483	0.3730
3.2513	20.6817	71000	3.5408	0.3737
3.2646	20.9730	72000	3.5328	0.3740
3.1964	21.2642	73000	3.5523	0.3730
3.2209	21.5555	74000	3.5462	0.3733
3.2431	21.8468	75000	3.5349	0.3740
3.1708	22.1381	76000	3.5538	0.3733
3.2133	22.4294	77000	3.5428	0.3736
3.2188	22.7207	78000	3.5397	0.3740
3.1532	23.0119	79000	3.5490	0.3735
3.1781	23.3033	80000	3.5497	0.3736
3.2226	23.5946	81000	3.5435	0.3740
3.2186	23.8859	82000	3.5341	0.3744
3.15	24.1771	83000	3.5553	0.3732
3.1901	24.4684	84000	3.5481	0.3741
3.2069	24.7597	85000	3.5377	0.3743
3.125	25.0510	86000	3.5503	0.3738
3.1808	25.3423	87000	3.5512	0.3739
3.1951	25.6336	88000	3.5415	0.3742
3.2001	25.9249	89000	3.5327	0.3747
3.14	26.2162	90000	3.5526	0.3738
3.1702	26.5075	91000	3.5467	0.3742
3.1881	26.7988	92000	3.5360	0.3748
3.118	27.0900	93000	3.5548	0.3738
3.1445	27.3813	94000	3.5487	0.3741
3.1757	27.6726	95000	3.5402	0.3746
3.1963	27.9639	96000	3.5341	0.3751
3.1336	28.2552	97000	3.5514	0.3741
3.1754	28.5465	98000	3.5453	0.3746
3.1774	28.8378	99000	3.5393	0.3749
3.1134	29.1290	100000	3.5526	0.3743
3.1463	29.4204	101000	3.5483	0.3746
3.1666	29.7117	102000	3.5405	0.3749
3.1574	30.0029	103000	3.5517	0.3743
3.1275	30.2942	104000	3.5561	0.3742
3.1505	30.5855	105000	3.5447	0.3749
3.1663	30.8768	106000	3.5387	0.3750
3.0979	31.1681	107000	3.5543	0.3745
3.1313	31.4594	108000	3.5532	0.3745
3.142	31.7507	109000	3.5411	0.3752

Framework versions

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