exceptions_exp2_swap_0.7_resemble_to_push_5039

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

• Loss: 3.5774
• Accuracy: 0.3665

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
4.8427	0.2915	1000	0.2535	4.7571
4.3511	0.5831	2000	0.2986	4.2858
4.1523	0.8746	3000	0.3146	4.1032
3.9998	1.1662	4000	0.3247	3.9920
3.9337	1.4577	5000	0.3309	3.9178
3.8708	1.7493	6000	0.3364	3.8585
3.7474	2.0408	7000	0.3409	3.8182
3.7592	2.3324	8000	0.3439	3.7848
3.7338	2.6239	9000	0.3464	3.7566
3.722	2.9155	10000	0.3487	3.7311
3.6432	3.2070	11000	0.3509	3.7173
3.6514	3.4985	12000	0.3525	3.6981
3.6411	3.7901	13000	0.3540	3.6814
3.5514	4.0816	14000	0.3552	3.6744
3.564	4.3732	15000	0.3563	3.6633
3.5733	4.6647	16000	0.3578	3.6498
3.5731	4.9563	17000	0.3591	3.6355
3.5143	5.2478	18000	0.3594	3.6401
3.5218	5.5394	19000	0.3602	3.6283
3.528	5.8309	20000	0.3613	3.6179
3.446	6.1224	21000	0.3616	3.6209
3.479	6.4140	22000	0.3623	3.6132
3.4919	6.7055	23000	0.3632	3.6044
3.505	6.9971	24000	0.3640	3.5939
3.4421	7.2886	25000	0.3637	3.6037
3.4527	7.5802	26000	0.3649	3.5910
3.4547	7.8717	27000	0.3653	3.5841
3.3729	8.1633	28000	0.3654	3.5948
3.4161	8.4548	29000	0.3656	3.5897
3.4295	8.7464	30000	0.3665	3.5774
3.3256	9.0379	31000	0.3664	3.5829
3.3733	9.3294	32000	0.3665	3.5812
3.4062	9.6210	33000	0.3673	3.5757
3.4021	9.9125	34000	0.3679	3.5638
3.3341	10.2041	35000	0.3671	3.5801
3.3611	10.4956	36000	0.3681	3.5724
3.3845	10.7872	37000	0.3686	3.5626
3.2814	11.0787	38000	0.3682	3.5742
3.3361	11.3703	39000	0.3686	3.5683
3.3702	11.6618	40000	0.3689	3.5626
3.3664	11.9534	41000	0.3696	3.5541
3.3073	12.2449	42000	0.3688	3.5696
3.3408	12.5364	43000	0.3694	3.5584
3.3521	12.8280	44000	0.3700	3.5495
3.2711	13.1195	45000	0.3692	3.5658
3.303	13.4111	46000	0.3699	3.5605
3.3211	13.7026	47000	0.3705	3.5522
3.3581	13.9942	48000	0.3713	3.5415
3.2851	14.2857	49000	0.3701	3.5600
3.3056	14.5773	50000	0.3707	3.5523
3.3251	14.8688	51000	0.3712	3.5443
3.2478	15.1603	52000	0.3708	3.5590
3.2882	15.4519	53000	0.3709	3.5550
3.3035	15.7434	54000	0.3714	3.5470
3.1955	16.0350	55000	0.3711	3.5545
3.2497	16.3265	56000	0.3708	3.5552
3.2795	16.6181	57000	0.3719	3.5448
3.2888	16.9096	58000	0.3724	3.5400
3.2231	17.2012	59000	0.3713	3.5586
3.2615	17.4927	60000	0.3716	3.5488
3.272	17.7843	61000	0.3722	3.5419
3.203	18.0758	62000	0.3719	3.5571
3.2376	18.3673	63000	0.3718	3.5523
3.2659	18.6589	64000	0.3725	3.5434
3.2758	18.9504	65000	0.3731	3.5348
3.2112	19.2420	66000	0.3721	3.5536
3.2374	19.5335	67000	0.3724	3.5478
3.2682	19.8251	68000	0.3730	3.5368
3.1777	20.1166	69000	0.3723	3.5554
3.227	20.4082	70000	0.3725	3.5494
3.2344	20.6997	71000	0.3728	3.5448
3.2637	20.9913	72000	0.3734	3.5349
3.206	21.2828	73000	0.3722	3.5538
3.227	21.5743	74000	0.3730	3.5461
3.2301	21.8659	75000	0.3735	3.5380
3.1727	22.1574	76000	0.3726	3.5528
3.2089	22.4490	77000	0.3728	3.5468
3.2303	22.7405	78000	0.3737	3.5382
3.1451	23.0321	79000	0.3730	3.5514
3.1934	23.3236	80000	0.3729	3.5496
3.1733	23.6152	81000	3.5532	0.3728
3.2024	23.9067	82000	3.5442	0.3732
3.1588	24.1983	83000	3.5548	0.3729
3.1918	24.4898	84000	3.5488	0.3731
3.204	24.7813	85000	3.5393	0.3737
3.1397	25.0729	86000	3.5539	0.3731
3.1854	25.3644	87000	3.5502	0.3733
3.1857	25.6560	88000	3.5422	0.3738
3.2093	25.9475	89000	3.5336	0.3743
3.1503	26.2391	90000	3.5533	0.3733
3.1896	26.5306	91000	3.5460	0.3738
3.182	26.8222	92000	3.5364	0.3744
3.132	27.1137	93000	3.5547	0.3734
3.1532	27.4052	94000	3.5490	0.3740
3.183	27.6968	95000	3.5422	0.3742
3.1996	27.9883	96000	3.5363	0.3744
3.1371	28.2799	97000	3.5531	0.3735
3.1674	28.5714	98000	3.5462	0.3740
3.1818	28.8630	99000	3.5386	0.3747
3.1052	29.1545	100000	3.5551	0.3736
3.1293	29.4461	101000	3.5497	0.3740
3.1532	29.7376	102000	3.5396	0.3744
3.0842	30.0292	103000	3.5559	0.3739
3.1202	30.3207	104000	3.5529	0.3739
3.1488	30.6122	105000	3.5447	0.3744
3.1591	30.9038	106000	3.5384	0.3750
3.1026	31.1953	107000	3.5590	0.3739
3.1355	31.4869	108000	3.5502	0.3741
3.1469	31.7784	109000	3.5438	0.3745

Framework versions

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