Nahuatl_Espanol_v2

This model is a fine-tuned version of google/flan-t5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.9402
• Bleu: 6.2508
• Gen Len: 50.5536

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: 3e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
No log	0.1064	100	3.0336	0.8102	55.6876
No log	0.2128	200	2.8627	0.9661	53.614
No log	0.3191	300	2.7696	1.111	53.6904
No log	0.4255	400	2.6947	1.1714	54.0762
3.1672	0.5319	500	2.6405	1.2824	53.5969
3.1672	0.6383	600	2.5967	1.3386	54.4867
3.1672	0.7447	700	2.5557	1.4915	55.2298
3.1672	0.8511	800	2.5261	1.5893	55.9804
3.1672	0.9574	900	2.4952	1.654	57.1207
2.8149	1.0638	1000	2.4734	1.7442	55.0846
2.8149	1.1702	1100	2.4484	1.8547	58.1569
2.8149	1.2766	1200	2.4287	1.9455	55.4888
2.8149	1.3830	1300	2.4103	2.0445	55.8386
2.8149	1.4894	1400	2.3908	2.2811	54.4788
2.6669	1.5957	1500	2.3750	2.4738	56.8398
2.6669	1.7021	1600	2.3572	2.5497	55.0454
2.6669	1.8085	1700	2.3422	2.7111	54.0798
2.6669	1.9149	1800	2.3286	2.8169	55.7837
2.6669	2.0213	1900	2.3147	2.9554	55.3014
2.5801	2.1277	2000	2.3018	3.133	54.3346
2.5801	2.2340	2100	2.2902	3.2281	55.0323
2.5801	2.3404	2200	2.2838	3.2981	56.7257
2.5801	2.4468	2300	2.2696	3.4102	54.1903
2.5801	2.5532	2400	2.2585	3.3897	55.325
2.5044	2.6596	2500	2.2480	3.6232	55.6974
2.5044	2.7660	2600	2.2401	3.6573	55.2994
2.5044	2.8723	2700	2.2306	3.722	56.7022
2.5044	2.9787	2800	2.2230	3.7379	52.895
2.5044	3.0851	2900	2.2132	3.7066	54.8602
2.4485	3.1915	3000	2.2064	3.9008	55.416
2.4485	3.2979	3100	2.1977	3.8825	54.9111
2.4485	3.4043	3200	2.1895	3.9786	54.3261
2.4485	3.5106	3300	2.1844	3.9746	54.5299
2.4485	3.6170	3400	2.1765	4.0218	55.0695
2.3988	3.7234	3500	2.1679	4.0382	56.6191
2.3988	3.8298	3600	2.1643	4.0658	54.9788
2.3988	3.9362	3700	2.1584	4.0867	54.61
2.3988	4.0426	3800	2.1540	4.3096	54.9816
2.3988	4.1489	3900	2.1455	4.2104	54.6118
2.3646	4.2553	4000	2.1413	4.4737	54.0416
2.3646	4.3617	4100	2.1350	4.4082	55.2328
2.3646	4.4681	4200	2.1300	4.3824	55.6597
2.3646	4.5745	4300	2.1252	4.4839	53.1048
2.3646	4.6809	4400	2.1185	4.5227	54.9721
2.3419	4.7872	4500	2.1130	4.3608	54.6448
2.3419	4.8936	4600	2.1119	4.5737	53.6723
2.3419	5.0	4700	2.1053	4.6235	53.8272
2.3419	5.1064	4800	2.0997	4.5814	53.8788
2.3419	5.2128	4900	2.0955	4.7139	53.5962
2.2982	5.3191	5000	2.0901	4.6879	53.3208
2.2982	5.4255	5100	2.0876	4.7353	53.6727
2.2982	5.5319	5200	2.0796	4.8038	53.7201
2.2982	5.6383	5300	2.0803	4.7483	53.5483
2.2982	5.7447	5400	2.0730	4.7057	53.3165
2.2785	5.8511	5500	2.0700	4.806	52.9666
2.2785	5.9574	5600	2.0679	4.9122	53.3892
2.2785	6.0638	5700	2.0642	4.9269	52.246
2.2785	6.1702	5800	2.0619	4.9346	52.926
2.2785	6.2766	5900	2.0560	5.1039	53.1269
2.2496	6.3830	6000	2.0550	5.1386	53.2045
2.2496	6.4894	6100	2.0504	5.2122	52.5518
2.2496	6.5957	6200	2.0460	5.1658	53.8375
2.2496	6.7021	6300	2.0441	5.2456	53.3426
2.2496	6.8085	6400	2.0399	5.2046	52.6617
2.2291	6.9149	6500	2.0359	5.1886	53.0398
2.2291	7.0213	6600	2.0342	5.3257	51.6602
2.2291	7.1277	6700	2.0323	5.2897	53.2622
2.2291	7.2340	6800	2.0298	5.4175	52.2951
2.2291	7.3404	6900	2.0271	5.4847	51.9924
2.2072	7.4468	7000	2.0240	5.4262	52.9876
2.2072	7.5532	7100	2.0205	5.5376	52.325
2.2072	7.6596	7200	2.0176	5.4789	52.4324
2.2072	7.7660	7300	2.0144	5.4898	52.2098
2.2072	7.8723	7400	2.0117	5.4634	52.3385
2.1996	7.9787	7500	2.0098	5.4655	52.7998
2.1996	8.0851	7600	2.0105	5.5251	52.1311
2.1996	8.1915	7700	2.0060	5.6941	51.5917
2.1996	8.2979	7800	2.0066	5.6255	52.1727
2.1996	8.4043	7900	2.0011	5.605	52.4629
2.172	8.5106	8000	2.0009	5.6421	51.6606
2.172	8.6170	8100	1.9979	5.7238	51.2952
2.172	8.7234	8200	1.9957	5.6869	51.3821
2.172	8.8298	8300	1.9924	5.7112	51.0052
2.172	8.9362	8400	1.9900	5.7394	51.8168
2.1697	9.0426	8500	1.9923	5.8348	51.0765
2.1697	9.1489	8600	1.9854	5.7641	51.7404
2.1697	9.2553	8700	1.9860	5.8078	50.6541
2.1697	9.3617	8800	1.9841	5.7624	51.7386
2.1697	9.4681	8900	1.9826	5.8623	51.401
2.1488	9.5745	9000	1.9814	5.887	50.9682
2.1488	9.6809	9100	1.9793	5.8872	50.88
2.1488	9.7872	9200	1.9777	5.8794	50.9482
2.1488	9.8936	9300	1.9742	5.8443	51.1684
2.1488	10.0	9400	1.9759	5.9447	51.2332
2.1508	10.1064	9500	1.9735	5.9591	51.3292
2.1508	10.2128	9600	1.9717	5.9751	51.5011
2.1508	10.3191	9700	1.9700	5.9655	50.8294
2.1508	10.4255	9800	1.9689	6.011	51.0793
2.1508	10.5319	9900	1.9683	5.9508	51.3352
2.1312	10.6383	10000	1.9658	5.9563	51.2867
2.1312	10.7447	10100	1.9635	5.9983	51.4218
2.1312	10.8511	10200	1.9616	6.0576	50.6682
2.1312	10.9574	10300	1.9618	6.0675	50.7527
2.1312	11.0638	10400	1.9604	6.1017	51.0262
2.1182	11.1702	10500	1.9603	6.114	50.9301
2.1182	11.2766	10600	1.9587	6.1085	51.0076
2.1182	11.3830	10700	1.9571	6.1066	51.0695
2.1182	11.4894	10800	1.9562	6.0495	51.5161
2.1182	11.5957	10900	1.9545	6.0907	50.8989
2.1194	11.7021	11000	1.9541	6.0534	50.7665
2.1194	11.8085	11100	1.9549	6.1778	50.403
2.1194	11.9149	11200	1.9528	6.1294	50.8481
2.1194	12.0213	11300	1.9510	6.1648	50.5486
2.1194	12.1277	11400	1.9526	6.1964	50.7805
2.1119	12.2340	11500	1.9506	6.1739	50.8039
2.1119	12.3404	11600	1.9502	6.1606	50.7453
2.1119	12.4468	11700	1.9490	6.2117	50.6436
2.1119	12.5532	11800	1.9485	6.1857	50.5681
2.1119	12.6596	11900	1.9471	6.1786	50.5037
2.0983	12.7660	12000	1.9470	6.1598	50.8716
2.0983	12.8723	12100	1.9453	6.174	50.8151
2.0983	12.9787	12200	1.9471	6.2005	50.6052
2.0983	13.0851	12300	1.9446	6.1764	50.6152
2.0983	13.1915	12400	1.9439	6.2014	50.8932
2.1012	13.2979	12500	1.9439	6.2146	50.7171
2.1012	13.4043	12600	1.9429	6.2222	50.6078
2.1012	13.5106	12700	1.9427	6.1982	50.7399
2.1012	13.6170	12800	1.9420	6.2085	50.8413
2.1012	13.7234	12900	1.9421	6.2133	50.6482
2.0958	13.8298	13000	1.9430	6.2267	50.6948
2.0958	13.9362	13100	1.9418	6.2637	50.5335
2.0958	14.0426	13200	1.9410	6.2697	50.5071
2.0958	14.1489	13300	1.9416	6.2494	50.5313
2.0958	14.2553	13400	1.9413	6.2439	50.5995
2.0922	14.3617	13500	1.9407	6.2484	50.509
2.0922	14.4681	13600	1.9407	6.2464	50.5193
2.0922	14.5745	13700	1.9403	6.2474	50.5404
2.0922	14.6809	13800	1.9405	6.2663	50.5403
2.0922	14.7872	13900	1.9403	6.26	50.5487
2.0898	14.8936	14000	1.9402	6.2518	50.5451
2.0898	15.0	14100	1.9402	6.2508	50.5536

Framework versions

• Transformers 4.40.2
• Pytorch 2.1.0
• Datasets 2.19.1
• Tokenizers 0.19.1