t5-small-entailement-Writer

This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5958

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: 2e-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: 150
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	42	1.8511
No log	2.0	84	1.2249
No log	3.0	126	0.9976
No log	4.0	168	0.9108
No log	5.0	210	0.8478
No log	6.0	252	0.8186
No log	7.0	294	0.7965
No log	8.0	336	0.7815
No log	9.0	378	0.7634
No log	10.0	420	0.7544
No log	11.0	462	0.7408
1.2198	12.0	504	0.7298
1.2198	13.0	546	0.7240
1.2198	14.0	588	0.7139
1.2198	15.0	630	0.7070
1.2198	16.0	672	0.7028
1.2198	17.0	714	0.6977
1.2198	18.0	756	0.6926
1.2198	19.0	798	0.6906
1.2198	20.0	840	0.6846
1.2198	21.0	882	0.6822
1.2198	22.0	924	0.6760
1.2198	23.0	966	0.6710
0.7403	24.0	1008	0.6667
0.7403	25.0	1050	0.6657
0.7403	26.0	1092	0.6653
0.7403	27.0	1134	0.6588
0.7403	28.0	1176	0.6584
0.7403	29.0	1218	0.6573
0.7403	30.0	1260	0.6520
0.7403	31.0	1302	0.6522
0.7403	32.0	1344	0.6525
0.7403	33.0	1386	0.6463
0.7403	34.0	1428	0.6453
0.7403	35.0	1470	0.6437
0.6642	36.0	1512	0.6397
0.6642	37.0	1554	0.6382
0.6642	38.0	1596	0.6365
0.6642	39.0	1638	0.6332
0.6642	40.0	1680	0.6335
0.6642	41.0	1722	0.6325
0.6642	42.0	1764	0.6295
0.6642	43.0	1806	0.6304
0.6642	44.0	1848	0.6287
0.6642	45.0	1890	0.6272
0.6642	46.0	1932	0.6267
0.6642	47.0	1974	0.6242
0.6127	48.0	2016	0.6232
0.6127	49.0	2058	0.6225
0.6127	50.0	2100	0.6211
0.6127	51.0	2142	0.6204
0.6127	52.0	2184	0.6196
0.6127	53.0	2226	0.6183
0.6127	54.0	2268	0.6168
0.6127	55.0	2310	0.6175
0.6127	56.0	2352	0.6160
0.6127	57.0	2394	0.6154
0.6127	58.0	2436	0.6143
0.6127	59.0	2478	0.6142
0.5799	60.0	2520	0.6131
0.5799	61.0	2562	0.6122
0.5799	62.0	2604	0.6120
0.5799	63.0	2646	0.6115
0.5799	64.0	2688	0.6119
0.5799	65.0	2730	0.6112
0.5799	66.0	2772	0.6099
0.5799	67.0	2814	0.6094
0.5799	68.0	2856	0.6082
0.5799	69.0	2898	0.6092
0.5799	70.0	2940	0.6081
0.5799	71.0	2982	0.6071
0.5558	72.0	3024	0.6062
0.5558	73.0	3066	0.6079
0.5558	74.0	3108	0.6072
0.5558	75.0	3150	0.6052
0.5558	76.0	3192	0.6066
0.5558	77.0	3234	0.6049
0.5558	78.0	3276	0.6042
0.5558	79.0	3318	0.6039
0.5558	80.0	3360	0.6050
0.5558	81.0	3402	0.6042
0.5558	82.0	3444	0.6040
0.5558	83.0	3486	0.6029
0.5292	84.0	3528	0.6032
0.5292	85.0	3570	0.6039
0.5292	86.0	3612	0.6036
0.5292	87.0	3654	0.6019
0.5292	88.0	3696	0.6014
0.5292	89.0	3738	0.6022
0.5292	90.0	3780	0.6014
0.5292	91.0	3822	0.6020
0.5292	92.0	3864	0.6028
0.5292	93.0	3906	0.5994
0.5292	94.0	3948	0.6004
0.5292	95.0	3990	0.5987
0.5159	96.0	4032	0.5992
0.5159	97.0	4074	0.5993
0.5159	98.0	4116	0.5989
0.5159	99.0	4158	0.6004
0.5159	100.0	4200	0.6001
0.5159	101.0	4242	0.6008
0.5159	102.0	4284	0.6006
0.5159	103.0	4326	0.5999
0.5159	104.0	4368	0.5994
0.5159	105.0	4410	0.5996
0.5159	106.0	4452	0.5991
0.5159	107.0	4494	0.5990
0.5004	108.0	4536	0.5996
0.5004	109.0	4578	0.5988
0.5004	110.0	4620	0.5992
0.5004	111.0	4662	0.5984
0.5004	112.0	4704	0.5982
0.5004	113.0	4746	0.5973
0.5004	114.0	4788	0.5984
0.5004	115.0	4830	0.5973
0.5004	116.0	4872	0.5977
0.5004	117.0	4914	0.5970
0.5004	118.0	4956	0.5976
0.5004	119.0	4998	0.5962
0.488	120.0	5040	0.5969
0.488	121.0	5082	0.5965
0.488	122.0	5124	0.5969
0.488	123.0	5166	0.5972
0.488	124.0	5208	0.5966
0.488	125.0	5250	0.5962
0.488	126.0	5292	0.5966
0.488	127.0	5334	0.5960
0.488	128.0	5376	0.5969
0.488	129.0	5418	0.5960
0.488	130.0	5460	0.5960
0.483	131.0	5502	0.5960
0.483	132.0	5544	0.5965
0.483	133.0	5586	0.5965
0.483	134.0	5628	0.5963
0.483	135.0	5670	0.5965
0.483	136.0	5712	0.5962
0.483	137.0	5754	0.5963
0.483	138.0	5796	0.5961
0.483	139.0	5838	0.5963
0.483	140.0	5880	0.5964
0.483	141.0	5922	0.5957
0.483	142.0	5964	0.5957
0.4809	143.0	6006	0.5957
0.4809	144.0	6048	0.5956
0.4809	145.0	6090	0.5958
0.4809	146.0	6132	0.5958
0.4809	147.0	6174	0.5959
0.4809	148.0	6216	0.5958
0.4809	149.0	6258	0.5958
0.4809	150.0	6300	0.5958

Framework versions

• Transformers 4.25.1
• Pytorch 1.13.0+cu116
• Datasets 2.7.1
• Tokenizers 0.13.2