exceptions_exp2_swap_take_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: 3.5559
• Accuracy: 0.3699

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
4.8292	0.2911	1000	0.2552	4.7487
4.3376	0.5822	2000	0.2984	4.2872
4.1459	0.8733	3000	0.3157	4.0957
3.9905	1.1642	4000	0.3247	3.9943
3.9297	1.4553	5000	0.3321	3.9160
3.8713	1.7464	6000	0.3372	3.8573
3.749	2.0373	7000	0.3416	3.8153
3.7514	2.3284	8000	0.3443	3.7842
3.7353	2.6195	9000	0.3473	3.7533
3.7255	2.9106	10000	0.3498	3.7284
3.6365	3.2014	11000	0.3517	3.7148
3.6482	3.4925	12000	0.3538	3.6938
3.6418	3.7837	13000	0.3550	3.6775
3.5431	4.0745	14000	0.3564	3.6696
3.5659	4.3656	15000	0.3576	3.6569
3.5787	4.6567	16000	0.3588	3.6440
3.5848	4.9478	17000	0.3599	3.6322
3.5063	5.2387	18000	0.3605	3.6335
3.5229	5.5298	19000	0.3615	3.6234
3.5269	5.8209	20000	0.3624	3.6139
3.4495	6.1118	21000	0.3624	3.6174
3.4645	6.4029	22000	0.3634	3.6089
3.4803	6.6940	23000	0.3642	3.6005
3.4936	6.9851	24000	0.3648	3.5880
3.4136	7.2760	25000	0.3647	3.5985
3.4486	7.5671	26000	0.3652	3.5912
3.4626	7.8582	27000	0.3664	3.5788
3.3763	8.1490	28000	0.3664	3.5859
3.412	8.4401	29000	0.3667	3.5806
3.4214	8.7313	30000	0.3673	3.5731
3.3197	9.0221	31000	0.3674	3.5766
3.3643	9.3132	32000	0.3678	3.5777
3.3979	9.6043	33000	0.3683	3.5702
3.4203	9.8954	34000	0.3687	3.5602
3.3214	10.1863	35000	0.3683	3.5733
3.3611	10.4774	36000	0.3690	3.5655
3.3814	10.7685	37000	0.3696	3.5591
3.2789	11.0594	38000	0.3693	3.5675
3.338	11.3505	39000	0.3695	3.5645
3.3603	11.6416	40000	0.3699	3.5559
3.3678	11.9327	41000	0.3706	3.5506
3.2986	12.2236	42000	0.3698	3.5642
3.3294	12.5147	43000	0.3704	3.5575
3.3483	12.8058	44000	0.3709	3.5469
3.2553	13.0966	45000	0.3705	3.5626
3.297	13.3878	46000	0.3707	3.5610
3.33	13.6789	47000	0.3714	3.5497
3.3477	13.9700	48000	0.3718	3.5378
3.28	14.2608	49000	0.3710	3.5554
3.3168	14.5519	50000	0.3717	3.5458
3.327	14.8430	51000	0.3722	3.5407
3.2331	15.1339	52000	0.3718	3.5530
3.2849	15.4250	53000	0.3721	3.5461
3.2965	15.7161	54000	0.3725	3.5429
3.2504	16.0070	55000	0.3721	3.5494
3.2571	16.2981	56000	0.3722	3.5513
3.2734	16.5892	57000	0.3726	3.5427
3.2856	16.8803	58000	0.3732	3.5367
3.228	17.1712	59000	0.3724	3.5501
3.2498	17.4623	60000	0.3728	3.5440
3.2769	17.7534	61000	0.3731	3.5361
3.1885	18.0442	62000	0.3729	3.5492
3.2432	18.3354	63000	0.3725	3.5478
3.2447	18.6265	64000	0.3735	3.5401
3.275	18.9176	65000	0.3738	3.5305
3.2181	19.2084	66000	0.3732	3.5480
3.2447	19.4995	67000	0.3734	3.5428
3.2561	19.7906	68000	0.3739	3.5324
3.1663	20.0815	69000	0.3732	3.5481
3.2209	20.3726	70000	0.3736	3.5431
3.2481	20.6637	71000	0.3738	3.5360
3.2535	20.9548	72000	0.3743	3.5300
3.1894	21.2457	73000	0.3734	3.5476
3.2223	21.5368	74000	0.3738	3.5424
3.2304	21.8279	75000	0.3740	3.5370
3.168	22.1188	76000	0.3737	3.5490
3.1925	22.4099	77000	0.3742	3.5440
3.2226	22.7010	78000	0.3742	3.5366
3.2524	22.9921	79000	0.3749	3.5267
3.1901	23.2830	80000	0.3739	3.5462
3.1873	23.5741	81000	3.5477	0.3738
3.1987	23.8652	82000	3.5427	0.3741
3.1555	24.1563	83000	3.5512	0.3739
3.1858	24.4474	84000	3.5436	0.3744
3.2125	24.7385	85000	3.5382	0.3744
3.1166	25.0294	86000	3.5464	0.3744
3.1496	25.3205	87000	3.5486	0.3739
3.1868	25.6116	88000	3.5410	0.3748
3.1976	25.9027	89000	3.5317	0.3754
3.1476	26.1936	90000	3.5496	0.3741
3.1707	26.4847	91000	3.5401	0.3749
3.1878	26.7758	92000	3.5360	0.3748
3.0964	27.0667	93000	3.5488	0.3744
3.15	27.3578	94000	3.5456	0.3745
3.1634	27.6489	95000	3.5373	0.3751
3.1898	27.9400	96000	3.5341	0.3752
3.1251	28.2308	97000	3.5520	0.3747
3.1432	28.5219	98000	3.5406	0.3750
3.1525	28.8131	99000	3.5348	0.3755
3.0951	29.1039	100000	3.5495	0.3747

Framework versions

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