exceptions_exp2_swap_0.7_resemble_to_carry_5039

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

• Loss: 3.5615
• Accuracy: 0.3689

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
4.8242	0.2915	1000	0.2546	4.7520
4.3466	0.5831	2000	0.2988	4.2851
4.1504	0.8746	3000	0.3153	4.0988
4.0097	1.1662	4000	0.3250	3.9910
3.9354	1.4577	5000	0.3308	3.9181
3.8851	1.7493	6000	0.3365	3.8563
3.7527	2.0408	7000	0.3407	3.8148
3.7615	2.3324	8000	0.3436	3.7850
3.745	2.6239	9000	0.3461	3.7589
3.7224	2.9155	10000	0.3489	3.7296
3.641	3.2070	11000	0.3508	3.7179
3.6476	3.4985	12000	0.3525	3.6981
3.6437	3.7901	13000	0.3540	3.6811
3.5453	4.0816	14000	0.3555	3.6750
3.5829	4.3732	15000	0.3566	3.6643
3.5717	4.6647	16000	0.3575	3.6523
3.5812	4.9563	17000	0.3589	3.6361
3.5058	5.2478	18000	0.3593	3.6392
3.5421	5.5394	19000	0.3602	3.6286
3.5335	5.8309	20000	0.3613	3.6166
3.4487	6.1224	21000	0.3613	3.6217
3.4802	6.4140	22000	0.3623	3.6133
3.4863	6.7055	23000	0.3629	3.6066
3.4923	6.9971	24000	0.3641	3.5928
3.4336	7.2886	25000	0.3638	3.6057
3.4596	7.5802	26000	0.3640	3.5962
3.4691	7.8717	27000	0.3652	3.5839
3.3796	8.1633	28000	0.3651	3.5963
3.42	8.4548	29000	0.3655	3.5873
3.4154	8.7464	30000	0.3662	3.5792
3.3327	9.0379	31000	0.3664	3.5865
3.371	9.3294	32000	0.3665	3.5838
3.3878	9.6210	33000	0.3669	3.5752
3.4186	9.9125	34000	0.3677	3.5661
3.3442	10.2041	35000	0.3669	3.5798
3.3714	10.4956	36000	0.3680	3.5713
3.3924	10.7872	37000	0.3681	3.5653
3.2878	11.0787	38000	0.3680	3.5735
3.3469	11.3703	39000	0.3684	3.5694
3.3636	11.6618	40000	0.3689	3.5615
3.3707	11.9534	41000	0.3690	3.5565
3.302	12.2449	42000	0.3688	3.5693
3.3389	12.5364	43000	0.3692	3.5596
3.347	12.8280	44000	0.3699	3.5556
3.2692	13.1195	45000	0.3692	3.5664
3.3144	13.4111	46000	0.3697	3.5613
3.3245	13.7026	47000	0.3703	3.5537
3.3386	13.9942	48000	0.3705	3.5437
3.2894	14.2857	49000	0.3700	3.5620
3.3067	14.5773	50000	0.3704	3.5531
3.3289	14.8688	51000	0.3710	3.5460
3.2443	15.1603	52000	0.3701	3.5632
3.2892	15.4519	53000	0.3708	3.5527
3.309	15.7434	54000	0.3712	3.5450
3.2073	16.0350	55000	0.3709	3.5567
3.2683	16.3265	56000	0.3712	3.5555
3.2914	16.6181	57000	0.3715	3.5488
3.2979	16.9096	58000	0.3719	3.5415
3.2297	17.2012	59000	0.3714	3.5575
3.2644	17.4927	60000	0.3715	3.5499
3.2676	17.7843	61000	0.3720	3.5435
3.1961	18.0758	62000	0.3716	3.5547
3.2453	18.3673	63000	0.3719	3.5521
3.2669	18.6589	64000	0.3722	3.5441
3.2748	18.9504	65000	0.3727	3.5391
3.2158	19.2420	66000	0.3716	3.5583
3.2428	19.5335	67000	0.3722	3.5464
3.2607	19.8251	68000	0.3726	3.5399
3.1795	20.1166	69000	0.3718	3.5516
3.2268	20.4082	70000	0.3717	3.5535
3.2474	20.6997	71000	0.3729	3.5401
3.2545	20.9913	72000	0.3732	3.5335
3.2048	21.2828	73000	0.3722	3.5539
3.2311	21.5743	74000	0.3727	3.5443
3.2426	21.8659	75000	0.3728	3.5400
3.1758	22.1574	76000	0.3724	3.5579
3.2092	22.4490	77000	0.3726	3.5501
3.2273	22.7405	78000	0.3732	3.5414
3.1406	23.0321	79000	0.3725	3.5562
3.191	23.3236	80000	0.3729	3.5503
3.1752	23.6152	81000	3.5545	0.3727
3.1974	23.9067	82000	3.5510	0.3728
3.1648	24.1983	83000	3.5604	0.3723
3.1822	24.4898	84000	3.5512	0.3730
3.2042	24.7813	85000	3.5416	0.3734
3.1414	25.0729	86000	3.5594	0.3728
3.1668	25.3644	87000	3.5537	0.3729
3.1958	25.6560	88000	3.5447	0.3736
3.2173	25.9475	89000	3.5381	0.3740
3.1565	26.2391	90000	3.5569	0.3731
3.1766	26.5306	91000	3.5484	0.3735
3.1968	26.8222	92000	3.5424	0.3739
3.1114	27.1137	93000	3.5616	0.3730
3.1604	27.4052	94000	3.5549	0.3734
3.1831	27.6968	95000	3.5406	0.3739
3.1993	27.9883	96000	3.5393	0.3743
3.1461	28.2799	97000	3.5545	0.3734
3.1724	28.5714	98000	3.5498	0.3738
3.1663	28.8630	99000	3.5400	0.3745
3.1213	29.1545	100000	3.5604	0.3733

Framework versions

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