exceptions_exp2_swap_require_to_push_40817

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

• Loss: 3.5524
• Accuracy: 0.3701

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: 40817
• 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
4.8279	0.2911	1000	0.2558	4.7466
4.343	0.5822	2000	0.2995	4.2822
4.1464	0.8733	3000	0.3150	4.0987
4.0028	1.1642	4000	0.3246	3.9904
3.9376	1.4553	5000	0.3322	3.9137
3.8855	1.7464	6000	0.3369	3.8564
3.7566	2.0373	7000	0.3414	3.8144
3.7449	2.3284	8000	0.3442	3.7841
3.7423	2.6195	9000	0.3472	3.7535
3.7271	2.9106	10000	0.3496	3.7284
3.6363	3.2014	11000	0.3513	3.7152
3.6469	3.4925	12000	0.3534	3.6948
3.6468	3.7837	13000	0.3549	3.6787
3.5401	4.0745	14000	0.3565	3.6696
3.5514	4.3656	15000	0.3574	3.6599
3.5807	4.6567	16000	0.3589	3.6468
3.5787	4.9478	17000	0.3597	3.6331
3.4973	5.2387	18000	0.3599	3.6367
3.5073	5.5298	19000	0.3612	3.6249
3.5316	5.8209	20000	0.3621	3.6134
3.4362	6.1118	21000	0.3626	3.6143
3.4731	6.4029	22000	0.3633	3.6092
3.4895	6.6940	23000	0.3640	3.5987
3.487	6.9851	24000	0.3651	3.5883
3.4367	7.2760	25000	0.3648	3.5982
3.4436	7.5671	26000	0.3653	3.5908
3.4507	7.8582	27000	0.3663	3.5801
3.395	8.1490	28000	0.3657	3.5901
3.3975	8.4401	29000	0.3664	3.5823
3.4255	8.7313	30000	0.3670	3.5760
3.3163	9.0221	31000	0.3674	3.5784
3.3875	9.3132	32000	0.3674	3.5773
3.3938	9.6043	33000	0.3682	3.5694
3.4113	9.8954	34000	0.3685	3.5590
3.3293	10.1863	35000	0.3681	3.5718
3.3613	10.4774	36000	0.3686	3.5668
3.3765	10.7685	37000	0.3691	3.5607
3.2937	11.0594	38000	0.3690	3.5686
3.3276	11.3505	39000	0.3695	3.5663
3.3714	11.6416	40000	0.3701	3.5524
3.364	11.9327	41000	0.3705	3.5484
3.3017	12.2236	42000	0.3699	3.5635
3.3351	12.5147	43000	0.3704	3.5532
3.3442	12.8058	44000	0.3708	3.5491
3.2818	13.0966	45000	0.3703	3.5614
3.2944	13.3878	46000	0.3709	3.5547
3.307	13.6789	47000	0.3713	3.5492
3.3449	13.9700	48000	0.3717	3.5390
3.2718	14.2608	49000	0.3707	3.5557
3.2979	14.5519	50000	0.3713	3.5505
3.3175	14.8430	51000	0.3720	3.5395
3.2373	15.1339	52000	0.3712	3.5549
3.2763	15.4250	53000	0.3714	3.5531
3.2976	15.7161	54000	0.3721	3.5427
3.2506	16.0070	55000	0.3716	3.5508
3.2402	16.2981	56000	0.3720	3.5513
3.2742	16.5892	57000	0.3723	3.5425
3.2783	16.8803	58000	0.3731	3.5359
3.2256	17.1712	59000	0.3723	3.5534
3.2532	17.4623	60000	0.3725	3.5479
3.267	17.7534	61000	0.3731	3.5368
3.1815	18.0442	62000	0.3725	3.5511
3.2283	18.3354	63000	0.3727	3.5452
3.2586	18.6265	64000	0.3731	3.5385
3.2732	18.9176	65000	0.3735	3.5357
3.1993	19.2084	66000	0.3727	3.5495
3.2264	19.4995	67000	0.3729	3.5461
3.2467	19.7906	68000	0.3738	3.5347
3.1678	20.0815	69000	0.3729	3.5486
3.2158	20.3726	70000	0.3729	3.5463
3.2199	20.6637	71000	0.3739	3.5360
3.2502	20.9548	72000	0.3743	3.5303
3.1779	21.2457	73000	0.3732	3.5511
3.2057	21.5368	74000	0.3734	3.5397
3.224	21.8279	75000	0.3740	3.5344
3.1673	22.1188	76000	0.3733	3.5502
3.193	22.4099	77000	0.3735	3.5465
3.2089	22.7010	78000	0.3740	3.5350
3.2209	22.9921	79000	0.3749	3.5294
3.1852	23.2830	80000	0.3737	3.5472
3.1845	23.5741	81000	3.5514	0.3737
3.2066	23.8652	82000	3.5411	0.3741
3.1408	24.1563	83000	3.5560	0.3732
3.1798	24.4474	84000	3.5463	0.3741
3.2038	24.7385	85000	3.5371	0.3745
3.1054	25.0294	86000	3.5532	0.3738
3.1681	25.3205	87000	3.5480	0.3740
3.1845	25.6116	88000	3.5392	0.3746
3.2036	25.9027	89000	3.5328	0.3747
3.1402	26.1936	90000	3.5533	0.3740
3.1603	26.4847	91000	3.5418	0.3746
3.1797	26.7758	92000	3.5376	0.3746
3.1215	27.0667	93000	3.5560	0.3742
3.1473	27.3578	94000	3.5487	0.3742
3.1778	27.6489	95000	3.5401	0.3748
3.1753	27.9400	96000	3.5342	0.3751
3.1257	28.2308	97000	3.5528	0.3741
3.16	28.5219	98000	3.5436	0.3749
3.171	28.8131	99000	3.5355	0.3752
3.086	29.1039	100000	3.5487	0.3747

Framework versions

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