exceptions_exp2_swap_require_to_carry_2128

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

• Loss: 3.5558
• Accuracy: 0.3698

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
4.8289	0.2911	1000	0.2545	4.7522
4.3295	0.5822	2000	0.2993	4.2868
4.1452	0.8733	3000	0.3157	4.0962
3.9889	1.1642	4000	0.3255	3.9910
3.9314	1.4553	5000	0.3320	3.9145
3.8805	1.7464	6000	0.3373	3.8562
3.7514	2.0373	7000	0.3416	3.8151
3.7624	2.3284	8000	0.3443	3.7867
3.7472	2.6195	9000	0.3473	3.7572
3.73	2.9106	10000	0.3497	3.7268
3.6305	3.2014	11000	0.3517	3.7154
3.6491	3.4925	12000	0.3532	3.6969
3.6344	3.7837	13000	0.3551	3.6789
3.5324	4.0745	14000	0.3564	3.6751
3.5684	4.3656	15000	0.3573	3.6606
3.5802	4.6567	16000	0.3585	3.6468
3.5646	4.9478	17000	0.3593	3.6362
3.5021	5.2387	18000	0.3602	3.6348
3.5145	5.5298	19000	0.3611	3.6240
3.5278	5.8209	20000	0.3621	3.6141
3.4424	6.1118	21000	0.3628	3.6201
3.4676	6.4029	22000	0.3636	3.6072
3.4864	6.6940	23000	0.3643	3.5997
3.4952	6.9851	24000	0.3649	3.5905
3.4255	7.2760	25000	0.3647	3.5982
3.4437	7.5671	26000	0.3657	3.5909
3.4696	7.8582	27000	0.3664	3.5787
3.3754	8.1490	28000	0.3660	3.5915
3.4158	8.4401	29000	0.3668	3.5868
3.4278	8.7313	30000	0.3674	3.5733
3.3285	9.0221	31000	0.3673	3.5804
3.3768	9.3132	32000	0.3675	3.5769
3.3918	9.6043	33000	0.3678	3.5706
3.4235	9.8954	34000	0.3686	3.5624
3.3301	10.1863	35000	0.3684	3.5713
3.3657	10.4774	36000	0.3688	3.5657
3.3882	10.7685	37000	0.3694	3.5586
3.2774	11.0594	38000	0.3694	3.5661
3.3418	11.3505	39000	0.3696	3.5666
3.3626	11.6416	40000	0.3697	3.5558
3.3769	11.9327	41000	0.3706	3.5470
3.3102	12.2236	42000	0.3700	3.5622
3.3311	12.5147	43000	0.3704	3.5521
3.3529	12.8058	44000	0.3708	3.5514
3.2645	13.0966	45000	0.3704	3.5610
3.3027	13.3878	46000	0.3708	3.5551
3.3317	13.6789	47000	0.3714	3.5486
3.3341	13.9700	48000	0.3716	3.5409
3.2718	14.2608	49000	0.3712	3.5574
3.3059	14.5519	50000	0.3718	3.5486
3.321	14.8430	51000	0.3722	3.5415
3.2297	15.1339	52000	0.3714	3.5563
3.2868	15.4250	53000	0.3716	3.5502
3.291	15.7161	54000	0.3723	3.5414
3.2615	16.0070	55000	0.3721	3.5496
3.2497	16.2981	56000	0.3722	3.5505
3.2737	16.5892	57000	0.3726	3.5432
3.3034	16.8803	58000	0.3731	3.5342
3.2227	17.1712	59000	0.3724	3.5486
3.2673	17.4623	60000	0.3726	3.5469
3.2939	17.7534	61000	0.3732	3.5381
3.1934	18.0442	62000	0.3728	3.5510
3.2315	18.3354	63000	0.3729	3.5466
3.2589	18.6265	64000	0.3729	3.5403
3.2753	18.9176	65000	0.3737	3.5333
3.2123	19.2084	66000	0.3729	3.5489
3.2533	19.4995	67000	0.3732	3.5407
3.2451	19.7906	68000	0.3735	3.5344
3.1742	20.0815	69000	0.3732	3.5475
3.213	20.3726	70000	0.3732	3.5459
3.238	20.6637	71000	0.3739	3.5350
3.2504	20.9548	72000	0.3745	3.5290
3.1916	21.2457	73000	0.3733	3.5467
3.2117	21.5368	74000	0.3740	3.5397
3.2279	21.8279	75000	0.3744	3.5343
3.1556	22.1188	76000	0.3735	3.5510
3.1982	22.4099	77000	0.3739	3.5434
3.2193	22.7010	78000	0.3742	3.5373
3.2446	22.9921	79000	0.3747	3.5279
3.1809	23.2830	80000	0.3741	3.5482
3.1964	23.5741	81000	0.3741	3.5391
3.2268	23.8652	82000	0.3748	3.5308
3.1506	24.1560	83000	0.3740	3.5501
3.18	24.4471	84000	0.3740	3.5447
3.2124	24.7382	85000	0.3747	3.5349
3.1181	25.0291	86000	0.3742	3.5465
3.1627	25.3202	87000	0.3742	3.5454
3.1809	25.6113	88000	0.3746	3.5398
3.2053	25.9024	89000	0.3753	3.5295
3.1337	26.1933	90000	0.3742	3.5487
3.1622	26.4844	91000	3.5508	0.3741
3.1757	26.7755	92000	3.5435	0.3746
3.1123	27.0667	93000	3.5498	0.3742
3.1531	27.3578	94000	3.5481	0.3745
3.1674	27.6489	95000	3.5386	0.3747
3.1848	27.9400	96000	3.5311	0.3757
3.1228	28.2308	97000	3.5527	0.3743
3.1465	28.5219	98000	3.5446	0.3751
3.167	28.8131	99000	3.5344	0.3754
3.1033	29.1039	100000	3.5509	0.3742
3.1225	29.3950	101000	3.5461	0.3747
3.1671	29.6861	102000	3.5416	0.3752
3.1634	29.9772	103000	3.5333	0.3758
3.1025	30.2681	104000	3.5496	0.3748
3.132	30.5592	105000	3.5431	0.3751
3.1561	30.8503	106000	3.5347	0.3758
3.0942	31.1412	107000	3.5555	0.3744
3.1232	31.4323	108000	3.5457	0.3753
3.1337	31.7234	109000	3.5392	0.3757
3.0579	32.0143	110000	3.5478	0.3752

Framework versions

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