exceptions_exp2_swap_0.3_resemble_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: 3.5693
• Accuracy: 0.3729

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
4.8254	0.2915	1000	0.2520	4.7814
4.3429	0.5830	2000	0.2989	4.2852
4.1495	0.8745	3000	0.3148	4.1004
3.9961	1.1659	4000	0.3242	3.9964
3.9501	1.4574	5000	0.3307	3.9212
3.8819	1.7488	6000	0.3358	3.8631
3.746	2.0402	7000	0.3404	3.8231
3.761	2.3317	8000	0.3434	3.7897
3.7421	2.6232	9000	0.3460	3.7618
3.7268	2.9147	10000	0.3484	3.7356
3.6436	3.2061	11000	0.3502	3.7209
3.6544	3.4976	12000	0.3521	3.7036
3.6477	3.7891	13000	0.3538	3.6851
3.5517	4.0805	14000	0.3549	3.6776
3.5819	4.3719	15000	0.3560	3.6682
3.5693	4.6634	16000	0.3573	3.6542
3.5879	4.9549	17000	0.3585	3.6396
3.4941	5.2463	18000	0.3587	3.6452
3.5351	5.5378	19000	0.3600	3.6341
3.5235	5.8293	20000	0.3613	3.6208
3.4477	6.1207	21000	0.3613	3.6251
3.4856	6.4122	22000	0.3621	3.6170
3.4815	6.7037	23000	0.3628	3.6062
3.5027	6.9952	24000	0.3635	3.5982
3.4276	7.2865	25000	0.3637	3.6062
3.4657	7.5780	26000	0.3640	3.5983
3.4597	7.8695	27000	0.3648	3.5885
3.3868	8.1609	28000	0.3647	3.5994
3.4322	8.4524	29000	0.3651	3.5921
3.4325	8.7439	30000	0.3658	3.5838
3.3427	9.0353	31000	0.3658	3.5906
3.3823	9.3268	32000	0.3661	3.5883
3.4155	9.6183	33000	0.3667	3.5786
3.4158	9.9098	34000	0.3674	3.5709
3.3389	10.2011	35000	0.3672	3.5799
3.3614	10.4926	36000	0.3672	3.5778
3.3803	10.7841	37000	0.3679	3.5677
3.289	11.0755	38000	0.3677	3.5783
3.3381	11.3670	39000	0.3676	3.5754
3.3723	11.6585	40000	0.3686	3.5656
3.374	11.9500	41000	0.3692	3.5579
3.311	12.2414	42000	0.3684	3.5747
3.35	12.5329	43000	0.3690	3.5670
3.3499	12.8243	44000	0.3697	3.5552
3.2678	13.1157	45000	0.3690	3.5724
3.3057	13.4072	46000	0.3692	3.5654
3.3354	13.6987	47000	0.3696	3.5596
3.3392	13.9902	48000	0.3703	3.5502
3.2776	14.2816	49000	0.3700	3.5625
3.3019	14.5731	50000	0.3702	3.5588
3.337	14.8646	51000	0.3706	3.5505
3.2535	15.1559	52000	0.3698	3.5664
3.2812	15.4474	53000	0.3705	3.5595
3.31	15.7389	54000	0.3708	3.5518
3.2068	16.0303	55000	0.3704	3.5636
3.2614	16.3218	56000	0.3705	3.5616
3.2864	16.6133	57000	0.3709	3.5545
3.3048	16.9048	58000	0.3716	3.5447
3.2405	17.1962	59000	0.3709	3.5626
3.2623	17.4877	60000	0.3711	3.5550
3.2825	17.7792	61000	0.3715	3.5491
3.2008	18.0705	62000	0.3714	3.5612
3.2395	18.3620	63000	0.3712	3.5590
3.2585	18.6535	64000	0.3716	3.5504
3.2839	18.9450	65000	0.3721	3.5440
3.2091	19.2364	66000	0.3713	3.5641
3.2486	19.5279	67000	0.3719	3.5526
3.2699	19.8194	68000	0.3724	3.5451
3.1871	20.1108	69000	0.3714	3.5659
3.2314	20.4023	70000	0.3718	3.5571
3.2536	20.6938	71000	0.3722	3.5493
3.2651	20.9853	72000	0.3728	3.5397
3.2028	21.2766	73000	0.3716	3.5608
3.228	21.5681	74000	0.3725	3.5489
3.2562	21.8596	75000	0.3726	3.5433
3.1677	22.1510	76000	0.3720	3.5633
3.2113	22.4425	77000	0.3722	3.5576
3.2368	22.7340	78000	0.3726	3.5471
3.1288	23.0254	79000	0.3721	3.5624
3.1875	23.3169	80000	0.3719	3.5603
3.1866	23.6083	81000	3.5600	0.3719
3.2025	23.8998	82000	3.5537	0.3727
3.1717	24.1915	83000	3.5624	0.3726
3.1968	24.4830	84000	3.5568	0.3726
3.2125	24.7745	85000	3.5489	0.3729
3.1323	25.0659	86000	3.5694	0.3721
3.1738	25.3574	87000	3.5588	0.3728
3.1948	25.6489	88000	3.5495	0.3730
3.2198	25.9404	89000	3.5441	0.3737
3.1477	26.2317	90000	3.5598	0.3727
3.188	26.5232	91000	3.5524	0.3730
3.1925	26.8147	92000	3.5479	0.3734
3.1235	27.1061	93000	3.5655	0.3726
3.1649	27.3976	94000	3.5588	0.3728
3.1772	27.6891	95000	3.5503	0.3733
3.2038	27.9806	96000	3.5452	0.3737
3.1323	28.2720	97000	3.5606	0.3729
3.1696	28.5635	98000	3.5562	0.3732
3.1698	28.8550	99000	3.5474	0.3738
3.1194	29.1463	100000	3.5693	0.3729

Framework versions

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