exceptions_exp2_swap_require_to_hit_5039

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

• Loss: 3.5570
• Accuracy: 0.3697

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	Accuracy	Validation Loss
4.8324	0.2911	1000	0.2543	4.7536
4.3358	0.5822	2000	0.2992	4.2848
4.1424	0.8733	3000	0.3156	4.0940
3.9889	1.1642	4000	0.3254	3.9893
3.9289	1.4553	5000	0.3320	3.9152
3.8718	1.7464	6000	0.3368	3.8573
3.7494	2.0373	7000	0.3414	3.8144
3.7529	2.3284	8000	0.3444	3.7842
3.7361	2.6195	9000	0.3471	3.7531
3.7257	2.9106	10000	0.3496	3.7297
3.639	3.2014	11000	0.3517	3.7152
3.649	3.4925	12000	0.3534	3.6955
3.6428	3.7837	13000	0.3546	3.6796
3.5444	4.0745	14000	0.3561	3.6706
3.5663	4.3656	15000	0.3570	3.6595
3.5791	4.6567	16000	0.3587	3.6446
3.5849	4.9478	17000	0.3599	3.6323
3.5071	5.2387	18000	0.3604	3.6357
3.5229	5.5298	19000	0.3612	3.6249
3.5281	5.8209	20000	0.3622	3.6134
3.449	6.1118	21000	0.3621	3.6178
3.466	6.4029	22000	0.3629	3.6095
3.4793	6.6940	23000	0.3641	3.5999
3.4936	6.9851	24000	0.3648	3.5884
3.4139	7.2760	25000	0.3647	3.5995
3.4495	7.5671	26000	0.3652	3.5907
3.4641	7.8582	27000	0.3662	3.5811
3.3763	8.1490	28000	0.3663	3.5884
3.4138	8.4401	29000	0.3666	3.5821
3.421	8.7313	30000	0.3672	3.5711
3.3207	9.0221	31000	0.3669	3.5782
3.3665	9.3132	32000	0.3676	3.5785
3.3992	9.6043	33000	0.3684	3.5692
3.4206	9.8954	34000	0.3686	3.5616
3.3227	10.1863	35000	0.3681	3.5742
3.3613	10.4774	36000	0.3690	3.5660
3.3809	10.7685	37000	0.3691	3.5599
3.2804	11.0594	38000	0.3692	3.5678
3.3386	11.3505	39000	0.3692	3.5646
3.3614	11.6416	40000	0.3697	3.5570
3.3704	11.9327	41000	0.3704	3.5498
3.3013	12.2236	42000	0.3698	3.5637
3.3299	12.5147	43000	0.3703	3.5568
3.3492	12.8058	44000	0.3709	3.5468
3.2567	13.0966	45000	0.3703	3.5606
3.2981	13.3878	46000	0.3707	3.5611
3.331	13.6789	47000	0.3713	3.5492
3.3485	13.9700	48000	0.3718	3.5411
3.2817	14.2608	49000	0.3710	3.5551
3.3184	14.5519	50000	0.3714	3.5471
3.3271	14.8430	51000	0.3719	3.5417
3.2353	15.1339	52000	0.3718	3.5549
3.2863	15.4250	53000	0.3721	3.5476
3.2966	15.7161	54000	0.3722	3.5417
3.2509	16.0070	55000	0.3720	3.5503
3.2588	16.2981	56000	0.3720	3.5524
3.2756	16.5892	57000	0.3725	3.5432
3.2862	16.8803	58000	0.3731	3.5356
3.2286	17.1712	59000	0.3722	3.5490
3.2498	17.4623	60000	0.3727	3.5422
3.2777	17.7534	61000	0.3731	3.5362
3.1902	18.0442	62000	0.3727	3.5498
3.2454	18.3354	63000	0.3725	3.5481
3.2458	18.6265	64000	0.3734	3.5383
3.2764	18.9176	65000	0.3737	3.5297
3.2186	19.2084	66000	0.3731	3.5486
3.2464	19.4995	67000	0.3733	3.5423
3.2576	19.7906	68000	0.3739	3.5338
3.1668	20.0815	69000	0.3729	3.5515
3.2226	20.3726	70000	0.3736	3.5436
3.2474	20.6637	71000	0.3737	3.5372
3.2568	20.9548	72000	0.3744	3.5301
3.1907	21.2457	73000	0.3736	3.5450
3.2237	21.5368	74000	0.3739	3.5394
3.2323	21.8279	75000	0.3741	3.5360
3.1697	22.1188	76000	0.3736	3.5491
3.1935	22.4099	77000	0.3740	3.5432
3.2237	22.7010	78000	0.3744	3.5365
3.2531	22.9921	79000	0.3745	3.5270
3.1898	23.2830	80000	0.3739	3.5471
3.2132	23.5741	81000	0.3743	3.5371
3.2254	23.8652	82000	0.3747	3.5309
3.1524	24.1560	83000	0.3741	3.5477
3.1867	24.4471	84000	0.3742	3.5420
3.2114	24.7382	85000	0.3746	3.5361
3.114	25.0291	86000	0.3742	3.5466
3.1505	25.3202	87000	0.3740	3.5459
3.1852	25.6113	88000	0.3749	3.5344
3.1963	25.9024	89000	0.3752	3.5335
3.1482	26.1933	90000	0.3739	3.5484
3.1526	26.4844	91000	3.5449	0.3743
3.1657	26.7755	92000	3.5409	0.3745
3.0996	27.0667	93000	3.5544	0.3741
3.1528	27.3578	94000	3.5471	0.3745
3.1649	27.6489	95000	3.5389	0.3748
3.1925	27.9400	96000	3.5311	0.3753
3.1249	28.2308	97000	3.5522	0.3748
3.1419	28.5219	98000	3.5402	0.3750
3.1525	28.8131	99000	3.5343	0.3752
3.094	29.1039	100000	3.5486	0.3746
3.138	29.3950	101000	3.5444	0.3747
3.1597	29.6861	102000	3.5398	0.3753
3.1702	29.9772	103000	3.5284	0.3757
3.12	30.2681	104000	3.5465	0.3746
3.1475	30.5592	105000	3.5391	0.3755
3.1588	30.8503	106000	3.5331	0.3756
3.0864	31.1412	107000	3.5506	0.3748
3.1193	31.4323	108000	3.5444	0.3750
3.1333	31.7234	109000	3.5376	0.3757
3.0675	32.0143	110000	3.5463	0.3753

Framework versions

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