exceptions_exp2_swap_0.7_resemble_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: 3.5637
• Accuracy: 0.3687

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	Accuracy	Validation Loss
4.846	0.2915	1000	0.2537	4.7586
4.3531	0.5831	2000	0.2980	4.2994
4.1598	0.8746	3000	0.3141	4.1056
4.0118	1.1662	4000	0.3238	4.0021
3.9368	1.4577	5000	0.3308	3.9226
3.8775	1.7493	6000	0.3356	3.8669
3.7361	2.0408	7000	0.3399	3.8211
3.7662	2.3324	8000	0.3429	3.7910
3.751	2.6239	9000	0.3457	3.7633
3.7261	2.9155	10000	0.3483	3.7345
3.6439	3.2070	11000	0.3499	3.7222
3.6571	3.4985	12000	0.3519	3.7034
3.6417	3.7901	13000	0.3538	3.6837
3.5507	4.0816	14000	0.3550	3.6785
3.5818	4.3732	15000	0.3559	3.6684
3.5894	4.6647	16000	0.3571	3.6542
3.5828	4.9563	17000	0.3583	3.6406
3.519	5.2478	18000	0.3589	3.6433
3.517	5.5394	19000	0.3599	3.6315
3.5413	5.8309	20000	0.3608	3.6203
3.4548	6.1224	21000	0.3612	3.6254
3.4783	6.4140	22000	0.3617	3.6183
3.4946	6.7055	23000	0.3629	3.6061
3.4963	6.9971	24000	0.3635	3.5969
3.4413	7.2886	25000	0.3634	3.6045
3.4646	7.5802	26000	0.3642	3.5956
3.4645	7.8717	27000	0.3648	3.5873
3.3822	8.1633	28000	0.3643	3.5992
3.4241	8.4548	29000	0.3653	3.5914
3.4308	8.7464	30000	0.3661	3.5819
3.3313	9.0379	31000	0.3657	3.5872
3.3749	9.3294	32000	0.3662	3.5830
3.4022	9.6210	33000	0.3667	3.5766
3.4153	9.9125	34000	0.3675	3.5685
3.3414	10.2041	35000	0.3668	3.5811
3.3628	10.4956	36000	0.3674	3.5757
3.3995	10.7872	37000	0.3679	3.5678
3.3102	11.0787	38000	0.3678	3.5756
3.3427	11.3703	39000	0.3682	3.5727
3.3409	11.6618	40000	0.3687	3.5637
3.3768	11.9534	41000	0.3692	3.5561
3.2986	12.2449	42000	0.3686	3.5683
3.3422	12.5364	43000	0.3691	3.5621
3.3658	12.8280	44000	0.3696	3.5554
3.2805	13.1195	45000	0.3691	3.5693
3.3208	13.4111	46000	0.3697	3.5644
3.3335	13.7026	47000	0.3700	3.5541
3.3366	13.9942	48000	0.3706	3.5455
3.2905	14.2857	49000	0.3696	3.5671
3.3022	14.5773	50000	0.3702	3.5574
3.3284	14.8688	51000	0.3708	3.5485
3.2506	15.1603	52000	0.3700	3.5641
3.2907	15.4519	53000	0.3705	3.5552
3.3118	15.7434	54000	0.3710	3.5500
3.2026	16.0350	55000	0.3703	3.5626
3.2675	16.3265	56000	0.3709	3.5573
3.2955	16.6181	57000	0.3713	3.5491
3.3104	16.9096	58000	0.3720	3.5415
3.2324	17.2012	59000	0.3711	3.5590
3.2573	17.4927	60000	0.3713	3.5533
3.283	17.7843	61000	0.3719	3.5437
3.2034	18.0758	62000	0.3714	3.5583
3.2529	18.3673	63000	0.3715	3.5578
3.2625	18.6589	64000	0.3719	3.5469
3.2832	18.9504	65000	0.3721	3.5428
3.2161	19.2420	66000	0.3712	3.5601
3.258	19.5335	67000	0.3721	3.5502
3.2695	19.8251	68000	0.3723	3.5441
3.1873	20.1166	69000	0.3713	3.5601
3.2182	20.4082	70000	0.3719	3.5562
3.2461	20.6997	71000	0.3723	3.5469
3.264	20.9913	72000	0.3729	3.5399
3.2139	21.2828	73000	0.3720	3.5547
3.2323	21.5743	74000	0.3723	3.5502
3.2557	21.8659	75000	0.3727	3.5446
3.17	22.1574	76000	0.3722	3.5586
3.2016	22.4490	77000	0.3726	3.5535
3.2393	22.7405	78000	0.3733	3.5440
3.1364	23.0321	79000	0.3726	3.5577
3.1812	23.3236	80000	0.3724	3.5569
3.1755	23.6152	81000	3.5591	0.3720
3.2106	23.9067	82000	3.5530	0.3730
3.1544	24.1983	83000	3.5606	0.3724
3.2014	24.4898	84000	3.5555	0.3727
3.2109	24.7813	85000	3.5434	0.3733
3.1416	25.0729	86000	3.5555	0.3728
3.1836	25.3644	87000	3.5540	0.3729
3.1971	25.6560	88000	3.5457	0.3735
3.2262	25.9475	89000	3.5401	0.3736
3.1439	26.2391	90000	3.5583	0.3726
3.1708	26.5306	91000	3.5483	0.3733
3.2046	26.8222	92000	3.5425	0.3741
3.1221	27.1137	93000	3.5582	0.3731
3.1633	27.4052	94000	3.5551	0.3732
3.1811	27.6968	95000	3.5441	0.3736
3.2128	27.9883	96000	3.5379	0.3741
3.1417	28.2799	97000	3.5575	0.3731
3.1628	28.5714	98000	3.5481	0.3737
3.1765	28.8630	99000	3.5427	0.3739
3.1272	29.1545	100000	3.5597	0.3731
3.157	29.4461	101000	3.5516	0.3739
3.1686	29.7376	102000	3.5437	0.3740
3.0863	30.0292	103000	3.5553	0.3736
3.136	30.3207	104000	3.5576	0.3735
3.145	30.6122	105000	3.5486	0.3738
3.176	30.9038	106000	3.5418	0.3741
3.1076	31.1953	107000	3.5598	0.3735
3.1439	31.4869	108000	3.5577	0.3735
3.1395	31.7784	109000	3.5455	0.3741
3.074	32.0700	110000	3.5608	0.3739
3.1258	32.3615	111000	3.5578	0.3738
3.1352	32.6531	112000	3.5467	0.3743
3.1551	32.9446	113000	3.5425	0.3746
3.0928	33.2362	114000	3.5655	0.3735
3.1325	33.5277	115000	3.5519	0.3740
3.1453	33.8192	116000	3.5490	0.3743

Framework versions

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