exceptions_exp2_swap_0.7_cost_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: 3.5813
• Accuracy: 0.3660

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
4.8495	0.2917	1000	4.7681	0.2531
4.3448	0.5834	2000	4.2971	0.2978
4.1576	0.8750	3000	4.1055	0.3141
4.0179	1.1665	4000	4.0037	0.3232
3.9363	1.4582	5000	3.9245	0.3301
3.8954	1.7499	6000	3.8654	0.3356
3.7561	2.0414	7000	3.8211	0.3404
3.7679	2.3331	8000	3.7919	0.3434
3.7508	2.6248	9000	3.7634	0.3460
3.7392	2.9165	10000	3.7330	0.3483
3.6454	3.2080	11000	3.7238	0.3501
3.6489	3.4996	12000	3.7063	0.3518
3.654	3.7913	13000	3.6838	0.3539
3.5456	4.0828	14000	3.6791	0.3552
3.5855	4.3745	15000	3.6678	0.3561
3.5876	4.6662	16000	3.6548	0.3573
3.5934	4.9579	17000	3.6400	0.3587
3.5238	5.2494	18000	3.6425	0.3591
3.5289	5.5411	19000	3.6325	0.3603
3.5296	5.8327	20000	3.6223	0.3610
3.4496	6.1243	21000	3.6246	0.3614
3.4715	6.4159	22000	3.6187	0.3620
3.5	6.7076	23000	3.6083	0.3626
3.4973	6.9993	24000	3.5957	0.3636
3.44	7.2908	25000	3.6067	0.3634
3.4574	7.5825	26000	3.5990	0.3643
3.4748	7.8742	27000	3.5894	0.3649
3.3905	8.1657	28000	3.5968	0.3648
3.4085	8.4574	29000	3.5889	0.3656
3.4401	8.7490	30000	3.5813	0.3660
3.3334	9.0405	31000	3.5882	0.3657
3.3827	9.3322	32000	3.5863	0.3661
3.4003	9.6239	33000	3.5744	0.3669
3.417	9.9156	34000	3.5681	0.3674
3.3425	10.2071	35000	3.5823	0.3671
3.389	10.4988	36000	3.5734	0.3675
3.4016	10.7905	37000	3.5658	0.3682
3.3073	11.0820	38000	3.5728	0.3680
3.3501	11.3736	39000	3.5740	0.3680
3.3746	11.6653	40000	3.5662	0.3688
3.3773	11.9570	41000	3.5572	0.3695
3.3141	12.2485	42000	3.5711	0.3684
3.3499	12.5402	43000	3.5616	0.3692
3.345	12.8319	44000	3.5553	0.3696
3.2648	13.1234	45000	3.5678	0.3693
3.3126	13.4151	46000	3.5607	0.3694
3.3294	13.7067	47000	3.5539	0.3703
3.3524	13.9984	48000	3.5477	0.3706
3.2961	14.2899	49000	3.5666	0.3698
3.3065	14.5816	50000	3.5565	0.3700
3.3238	14.8733	51000	3.5494	0.3708
3.2633	15.1648	52000	3.5653	0.3701
3.2915	15.4565	53000	3.5572	0.3710
3.3027	15.7482	54000	3.5493	0.3710
3.2114	16.0397	55000	3.5605	0.3707
3.2694	16.3313	56000	3.5598	0.3709
3.3015	16.6230	57000	3.5504	0.3713
3.2965	16.9147	58000	3.5434	0.3719
3.2518	17.2062	59000	3.5610	0.3710
3.2699	17.4979	60000	3.5537	0.3715
3.2833	17.7896	61000	3.5451	0.3720
3.205	18.0811	62000	3.5603	0.3713
3.2405	18.3728	63000	3.5543	0.3716
3.2622	18.6644	64000	3.5477	0.3721
3.2853	18.9561	65000	3.5399	0.3728
3.2136	19.2476	66000	3.5571	0.3717
3.2421	19.5393	67000	3.5518	0.3718
3.247	19.8310	68000	3.5424	0.3725
3.1915	20.1225	69000	3.5618	0.3720
3.2315	20.4142	70000	3.5545	0.3723
3.2387	20.7059	71000	3.5445	0.3727
3.2681	20.9975	72000	3.5392	0.3732
3.1985	21.2891	73000	3.5578	0.3719
3.2263	21.5807	74000	3.5477	0.3727
3.241	21.8724	75000	3.5404	0.3730
3.1623	22.1639	76000	3.5533	0.3726
3.2073	22.4556	77000	3.5501	0.3726
3.2213	22.7473	78000	3.5449	0.3731
3.1494	23.0388	79000	3.5543	0.3727
3.1892	23.3305	80000	3.5524	0.3726
3.211	23.6222	81000	3.5505	0.3730
3.2331	23.9138	82000	3.5388	0.3735
3.1621	24.2053	83000	3.5566	0.3725
3.2148	24.4970	84000	3.5493	0.3731
3.206	24.7887	85000	3.5413	0.3736
3.12	25.0802	86000	3.5574	0.3730
3.1795	25.3719	87000	3.5528	0.3731
3.1956	25.6636	88000	3.5424	0.3738
3.2272	25.9553	89000	3.5368	0.3739
3.1544	26.2468	90000	3.5519	0.3732
3.1801	26.5384	91000	3.5500	0.3734
3.2106	26.8301	92000	3.5401	0.3741
3.1296	27.1216	93000	3.5584	0.3732
3.1701	27.4133	94000	3.5518	0.3735
3.1742	27.7050	95000	3.5419	0.3741
3.1993	27.9967	96000	3.5368	0.3742
3.1572	28.2882	97000	3.5559	0.3732
3.169	28.5799	98000	3.5508	0.3736
3.189	28.8715	99000	3.5381	0.3743
3.1151	29.1630	100000	3.5557	0.3736
3.1515	29.4547	101000	3.5521	0.3739
3.1737	29.7464	102000	3.5451	0.3743
3.0872	30.0379	103000	3.5587	0.3737
3.1245	30.3296	104000	3.5546	0.3737
3.15	30.6213	105000	3.5496	0.3740
3.1683	30.9130	106000	3.5438	0.3744
3.0999	31.2045	107000	3.5577	0.3737
3.1229	31.4961	108000	3.5498	0.3743
3.1502	31.7878	109000	3.5434	0.3747

Framework versions

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