exceptions_exp2_swap_take_to_carry_3591

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

• Loss: 3.5486
• Accuracy: 0.3749

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: 3591
• 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
4.8419	0.2911	1000	0.2537	4.7597
4.334	0.5822	2000	0.3005	4.2763
4.1414	0.8733	3000	0.3162	4.0872
3.988	1.1642	4000	0.3262	3.9859
3.9291	1.4553	5000	0.3323	3.9104
3.8593	1.7464	6000	0.3378	3.8522
3.728	2.0373	7000	0.3424	3.8078
3.7387	2.3284	8000	0.3452	3.7773
3.726	2.6195	9000	0.3483	3.7500
3.7115	2.9106	10000	0.3504	3.7216
3.6265	3.2014	11000	0.3524	3.7080
3.6239	3.4925	12000	0.3538	3.6925
3.6338	3.7837	13000	0.3556	3.6747
3.5323	4.0745	14000	0.3568	3.6658
3.5517	4.3656	15000	0.3579	3.6559
3.5653	4.6567	16000	0.3591	3.6403
3.5804	4.9478	17000	0.3604	3.6273
3.4955	5.2387	18000	0.3609	3.6313
3.5186	5.5298	19000	0.3619	3.6206
3.528	5.8209	20000	0.3626	3.6092
3.4493	6.1121	21000	0.3625	3.6193
3.4741	6.4032	22000	0.3634	3.6079
3.4831	6.6943	23000	0.3642	3.5997
3.4884	6.9854	24000	0.3648	3.5914
3.4114	7.2763	25000	0.3652	3.5968
3.4409	7.5674	26000	0.3658	3.5891
3.4627	7.8585	27000	0.3664	3.5772
3.3681	8.1493	28000	0.3664	3.5901
3.4015	8.4404	29000	0.3668	3.5806
3.4324	8.7315	30000	0.3675	3.5704
3.3179	9.0224	31000	0.3675	3.5758
3.3711	9.3135	32000	0.3680	3.5750
3.3914	9.6046	33000	0.3684	3.5665
3.4059	9.8957	34000	0.3689	3.5596
3.3312	10.1866	35000	0.3683	3.5742
3.3566	10.4777	36000	0.3689	3.5626
3.3838	10.7688	37000	0.3696	3.5544
3.2859	11.0597	38000	0.3695	3.5634
3.3232	11.3508	39000	0.3695	3.5633
3.3579	11.6419	40000	0.3699	3.5554
3.359	11.9330	41000	0.3704	3.5463
3.2982	12.2239	42000	0.3700	3.5620
3.3201	12.5150	43000	0.3706	3.5555
3.3574	12.8061	44000	0.3711	3.5455
3.2629	13.0969	45000	0.3706	3.5631
3.2942	13.3880	46000	0.3707	3.5584
3.3158	13.6791	47000	0.3717	3.5460
3.3321	13.9702	48000	0.3719	3.5389
3.268	14.2611	49000	0.3712	3.5553
3.2944	14.5522	50000	0.3719	3.5450
3.3297	14.8433	51000	0.3722	3.5383
3.2321	15.1342	52000	0.3717	3.5537
3.2757	15.4253	53000	0.3715	3.5540
3.2903	15.7164	54000	0.3726	3.5403
3.2619	16.0073	55000	0.3722	3.5459
3.2347	16.2984	56000	0.3724	3.5482
3.2865	16.5895	57000	0.3726	3.5435
3.2945	16.8806	58000	0.3731	3.5342
3.2187	17.1715	59000	0.3721	3.5531
3.263	17.4626	60000	0.3726	3.5471
3.2745	17.7537	61000	0.3731	3.5349
3.1714	18.0445	62000	0.3727	3.5482
3.2333	18.3356	63000	0.3728	3.5471
3.2593	18.6267	64000	0.3733	3.5397
3.2765	18.9179	65000	0.3737	3.5319
3.1925	19.2087	66000	0.3728	3.5505
3.2296	19.4998	67000	0.3731	3.5457
3.2636	19.7909	68000	0.3738	3.5353
3.1597	20.0818	69000	0.3734	3.5464
3.2035	20.3729	70000	0.3737	3.5455
3.2408	20.6640	71000	0.3739	3.5365
3.2362	20.9551	72000	0.3743	3.5315
3.189	21.2460	73000	0.3732	3.5479
3.2198	21.5371	74000	0.3738	3.5381
3.2375	21.8282	75000	0.3744	3.5333
3.1701	22.1191	76000	0.3737	3.5474
3.194	22.4102	77000	0.3740	3.5464
3.2292	22.7013	78000	0.3743	3.5335
3.2351	22.9924	79000	0.3748	3.5284
3.1732	23.2832	80000	0.3736	3.5480
3.2045	23.5743	81000	0.3743	3.5388
3.208	23.8655	82000	0.3744	3.5338
3.148	24.1563	83000	0.3735	3.5505
3.1779	24.4474	84000	0.3744	3.5429
3.2029	24.7385	85000	0.3747	3.5331
3.1095	25.0294	86000	0.3742	3.5481
3.1525	25.3205	87000	0.3738	3.5492
3.1751	25.6116	88000	0.3743	3.5421
3.2118	25.9027	89000	0.3750	3.5319
3.1337	26.1936	90000	0.3742	3.5500
3.157	26.4844	91000	3.5515	0.3737
3.1701	26.7755	92000	3.5359	0.3748
3.1103	27.0667	93000	3.5545	0.3741
3.1519	27.3578	94000	3.5453	0.3744
3.1652	27.6489	95000	3.5364	0.3751
3.1913	27.9400	96000	3.5336	0.3752
3.1146	28.2308	97000	3.5465	0.3746
3.1591	28.5219	98000	3.5440	0.3748
3.1626	28.8131	99000	3.5396	0.3752
3.0864	29.1039	100000	3.5525	0.3744
3.1397	29.3950	101000	3.5431	0.3748
3.1457	29.6861	102000	3.5395	0.3755
3.174	29.9772	103000	3.5299	0.3756
3.1094	30.2681	104000	3.5471	0.3750
3.128	30.5592	105000	3.5418	0.3753
3.1526	30.8503	106000	3.5344	0.3757
3.0948	31.1412	107000	3.5509	0.3748
3.1264	31.4323	108000	3.5464	0.3750
3.1317	31.7234	109000	3.5374	0.3755
3.0649	32.0143	110000	3.5486	0.3749

Framework versions

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