learning_rate: 0.0006
train_batch_size: 16
eval_batch_size: 16
seed: 40817
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	Validation Loss	Accuracy
4.8312	0.2911	1000	4.7509	0.2552
4.3449	0.5822	2000	4.2835	0.2990
4.1477	0.8733	3000	4.1003	0.3147
4.0043	1.1642	4000	3.9900	0.3250
3.9387	1.4553	5000	3.9161	0.3320
3.8859	1.7464	6000	3.8558	0.3373
3.756	2.0373	7000	3.8135	0.3413
3.7452	2.3284	8000	3.7825	0.3446
3.7425	2.6195	9000	3.7513	0.3474
3.727	2.9106	10000	3.7274	0.3499
3.6362	3.2014	11000	3.7155	0.3515
3.6469	3.4925	12000	3.6953	0.3534
3.6474	3.7837	13000	3.6770	0.3550
3.5399	4.0745	14000	3.6720	0.3564
3.5524	4.3656	15000	3.6613	0.3574
3.5807	4.6567	16000	3.6470	0.3589
3.5784	4.9478	17000	3.6335	0.3600
3.4979	5.2387	18000	3.6358	0.3603
3.5083	5.5298	19000	3.6250	0.3615
3.5324	5.8209	20000	3.6133	0.3623
3.4379	6.1118	21000	3.6169	0.3627
3.4747	6.4029	22000	3.6101	0.3634
3.4903	6.6940	23000	3.6026	0.3637
3.49	6.9851	24000	3.5923	0.3648
3.4386	7.2760	25000	3.5994	0.3651
3.4432	7.5671	26000	3.5911	0.3656
3.4524	7.8582	27000	3.5819	0.3661
3.3966	8.1490	28000	3.5913	0.3657
3.3985	8.4401	29000	3.5819	0.3665
3.4273	8.7313	30000	3.5763	0.3670
3.3189	9.0221	31000	3.5819	0.3673
3.3906	9.3132	32000	3.5801	0.3674
3.3951	9.6043	33000	3.5708	0.3681
3.4121	9.8954	34000	3.5623	0.3685
3.3325	10.1863	35000	3.5728	0.3682
3.3645	10.4774	36000	3.5677	0.3686
3.3783	10.7685	37000	3.5611	0.3694
3.2953	11.0594	38000	3.5680	0.3692
3.3298	11.3505	39000	3.5662	0.3694
3.3732	11.6416	40000	3.5562	0.3698
3.3652	11.9327	41000	3.5494	0.3704
3.3035	12.2236	42000	3.5659	0.3699
3.3395	12.5147	43000	3.5551	0.3706
3.3462	12.8058	44000	3.5508	0.3709
3.2844	13.0966	45000	3.5628	0.3704
3.298	13.3878	46000	3.5562	0.3706
3.3092	13.6789	47000	3.5500	0.3712
3.3464	13.9700	48000	3.5427	0.3717
3.2749	14.2608	49000	3.5561	0.3707
3.2988	14.5519	50000	3.5517	0.3714
3.3202	14.8430	51000	3.5419	0.3718
3.2386	15.1339	52000	3.5554	0.3713
3.279	15.4250	53000	3.5537	0.3712
3.3002	15.7161	54000	3.5458	0.3719
3.2546	16.0070	55000	3.5516	0.3718
3.2415	16.2981	56000	3.5529	0.3717
3.2769	16.5892	57000	3.5461	0.3721
3.2811	16.8803	58000	3.5380	0.3731
3.2279	17.1712	59000	3.5542	0.3723
3.2545	17.4623	60000	3.5483	0.3725
3.2711	17.7534	61000	3.5383	0.3731
3.1835	18.0442	62000	3.5527	0.3726
3.229	18.3354	63000	3.5487	0.3727
3.2603	18.6265	64000	3.5396	0.3731
3.2764	18.9176	65000	3.5383	0.3735
3.2012	19.2084	66000	3.5521	0.3727
3.2295	19.4995	67000	3.5475	0.3729
3.2494	19.7906	68000	3.5368	0.3738
3.1692	20.0815	69000	3.5535	0.3729
3.2183	20.3726	70000	3.5476	0.3729
3.2231	20.6637	71000	3.5387	0.3737
3.2532	20.9548	72000	3.5326	0.3740
3.181	21.2457	73000	3.5512	0.3731
3.208	21.5368	74000	3.5410	0.3737
3.2263	21.8279	75000	3.5328	0.3739
3.1701	22.1188	76000	3.5513	0.3733
3.1964	22.4099	77000	3.5475	0.3733
3.2125	22.7010	78000	3.5367	0.3739
3.2233	22.9921	79000	3.5337	0.3745
3.1873	23.2830	80000	3.5494	0.3735
3.2054	23.5741	81000	3.5385	0.3745
3.2297	23.8652	82000	3.5337	0.3742
3.1424	24.1560	83000	3.5517	0.3737
3.1798	24.4471	84000	3.5448	0.3740
3.2046	24.7382	85000	3.5392	0.3745
3.1083	25.0291	86000	3.5515	0.3740
3.1664	25.3202	87000	3.5489	0.3742
3.1854	25.6113	88000	3.5397	0.3747
3.2033	25.9024	89000	3.5337	0.3748
3.1423	26.1933	90000	3.5548	0.3738
3.1645	26.4844	91000	3.5417	0.3745
3.1834	26.7755	92000	3.5382	0.3747

Framework versions

Transformers 4.55.2
Pytorch 2.8.0+cu128
Datasets 4.0.0
Tokenizers 0.21.4

craa/exceptions_exp2_swap_take_to_push_40817

exceptions_exp2_swap_take_to_push_40817

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

craa/
exceptions_exp2_swap_take_to_push_40817