msa_prot_t5_repr_seq

This model is a fine-tuned version of Rostlab/prot_t5_xl_uniref50 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.8396

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: 0.001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	10	2.9787
No log	2.0	20	2.9960
No log	3.0	30	2.9192
No log	4.0	40	2.9534
3.0706	5.0	50	2.9662
3.0706	6.0	60	2.9160
3.0706	7.0	70	2.9198
3.0706	8.0	80	2.9258
3.0706	9.0	90	2.8992
2.9097	10.0	100	2.8073
2.9097	11.0	110	2.8701
2.9097	12.0	120	2.8366
2.9097	13.0	130	2.7131
2.9097	14.0	140	2.7704
2.8396	15.0	150	2.9375
2.8396	16.0	160	2.7965
2.8396	17.0	170	2.7563
2.8396	18.0	180	2.8374
2.8396	19.0	190	2.7491
2.8057	20.0	200	2.6914
2.8057	21.0	210	2.7746
2.8057	22.0	220	2.8187
2.8057	23.0	230	2.9719
2.8057	24.0	240	2.8489
2.8127	25.0	250	2.8719
2.8127	26.0	260	2.8749
2.8127	27.0	270	2.7897
2.8127	28.0	280	2.8159
2.8127	29.0	290	2.8765
2.7912	30.0	300	2.7582
2.7912	31.0	310	2.7970
2.7912	32.0	320	2.8463
2.7912	33.0	330	2.8521
2.7912	34.0	340	2.7665
2.8258	35.0	350	2.7878
2.8258	36.0	360	2.8995
2.8258	37.0	370	3.0310
2.8258	38.0	380	2.9792
2.8258	39.0	390	2.8650
2.908	40.0	400	2.8697
2.908	41.0	410	2.9299
2.908	42.0	420	2.7992
2.908	43.0	430	2.9172
2.908	44.0	440	2.8923
2.8984	45.0	450	2.8248
2.8984	46.0	460	2.9112
2.8984	47.0	470	2.8829
2.8984	48.0	480	2.8336
2.8984	49.0	490	2.7418
2.8658	50.0	500	2.7437
2.8658	51.0	510	2.7814
2.8658	52.0	520	2.8369
2.8658	53.0	530	2.8406
2.8658	54.0	540	2.8157
2.8376	55.0	550	2.9553
2.8376	56.0	560	2.7017
2.8376	57.0	570	2.8666
2.8376	58.0	580	2.7793
2.8376	59.0	590	2.9166
2.8294	60.0	600	2.7619
2.8294	61.0	610	2.9795
2.8294	62.0	620	2.7319
2.8294	63.0	630	2.9738
2.8294	64.0	640	2.8191
2.8127	65.0	650	2.8016
2.8127	66.0	660	3.0365
2.8127	67.0	670	2.7354
2.8127	68.0	680	3.0375
2.8127	69.0	690	2.6959
2.8177	70.0	700	3.0138
2.8177	71.0	710	2.8042
2.8177	72.0	720	2.8472
2.8177	73.0	730	3.0400
2.8177	74.0	740	2.7783
2.7711	75.0	750	2.8213
2.7711	76.0	760	2.7525
2.7711	77.0	770	2.8102
2.7711	78.0	780	3.0207
2.7711	79.0	790	2.9376
2.7756	80.0	800	2.9294
2.7756	81.0	810	3.0247
2.7756	82.0	820	2.9156
2.7756	83.0	830	2.9402
2.7756	84.0	840	2.7519
2.7855	85.0	850	2.8340
2.7855	86.0	860	2.8383
2.7855	87.0	870	2.8201
2.7855	88.0	880	3.0234
2.7855	89.0	890	2.8864
2.7698	90.0	900	2.8733
2.7698	91.0	910	2.9433
2.7698	92.0	920	2.7214
2.7698	93.0	930	2.9910
2.7698	94.0	940	2.6898
2.7683	95.0	950	2.9439
2.7683	96.0	960	2.9992
2.7683	97.0	970	3.0757
2.7683	98.0	980	3.0063
2.7683	99.0	990	3.0445
2.7727	100.0	1000	2.8396

Framework versions

• Transformers 4.51.3
• Pytorch 2.7.0+cu126
• Datasets 3.5.0
• Tokenizers 0.21.1