best_model-sst-2-32-21

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

• Loss: 1.1726
• Accuracy: 0.8281

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: 1e-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
• lr_scheduler_warmup_steps: 500
• num_epochs: 150

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	2	1.3058	0.7812
No log	2.0	4	1.3056	0.7812
No log	3.0	6	1.3051	0.7812
No log	4.0	8	1.3041	0.7812
0.6356	5.0	10	1.3027	0.7812
0.6356	6.0	12	1.3014	0.7812
0.6356	7.0	14	1.3002	0.7812
0.6356	8.0	16	1.2983	0.7812
0.6356	9.0	18	1.2961	0.7812
0.4734	10.0	20	1.2940	0.7812
0.4734	11.0	22	1.2916	0.7812
0.4734	12.0	24	1.2888	0.7812
0.4734	13.0	26	1.2848	0.7812
0.4734	14.0	28	1.2796	0.7812
0.3467	15.0	30	1.2731	0.7812
0.3467	16.0	32	1.2665	0.7812
0.3467	17.0	34	1.2594	0.7812
0.3467	18.0	36	1.2530	0.7812
0.3467	19.0	38	1.2462	0.7969
0.2773	20.0	40	1.2393	0.7969
0.2773	21.0	42	1.2324	0.7969
0.2773	22.0	44	1.2262	0.7969
0.2773	23.0	46	1.2197	0.7969
0.2773	24.0	48	1.2104	0.7969
0.1546	25.0	50	1.1969	0.7969
0.1546	26.0	52	1.1807	0.7969
0.1546	27.0	54	1.1697	0.7969
0.1546	28.0	56	1.1534	0.7969
0.1546	29.0	58	1.1462	0.7969
0.1236	30.0	60	1.1457	0.7969
0.1236	31.0	62	1.1458	0.7969
0.1236	32.0	64	1.1435	0.7969
0.1236	33.0	66	1.1347	0.8125
0.1236	34.0	68	1.1284	0.8125
0.0234	35.0	70	1.1252	0.8125
0.0234	36.0	72	1.1232	0.8125
0.0234	37.0	74	1.1203	0.8125
0.0234	38.0	76	1.1165	0.8125
0.0234	39.0	78	1.1136	0.8125
0.0266	40.0	80	1.1115	0.8125
0.0266	41.0	82	1.1144	0.8125
0.0266	42.0	84	1.1234	0.8125
0.0266	43.0	86	1.1296	0.7969
0.0266	44.0	88	1.1241	0.8125
0.0068	45.0	90	1.1150	0.8125
0.0068	46.0	92	1.1060	0.8125
0.0068	47.0	94	1.0985	0.8125
0.0068	48.0	96	1.0928	0.8125
0.0068	49.0	98	1.0899	0.8125
0.0016	50.0	100	1.0867	0.8125
0.0016	51.0	102	1.0921	0.8125
0.0016	52.0	104	1.0992	0.8125
0.0016	53.0	106	1.1055	0.8125
0.0016	54.0	108	1.1111	0.8281
0.0014	55.0	110	1.1121	0.8125
0.0014	56.0	112	1.1059	0.8125
0.0014	57.0	114	1.0984	0.8125
0.0014	58.0	116	1.0892	0.8125
0.0014	59.0	118	1.0797	0.8125
0.0024	60.0	120	1.0721	0.8125
0.0024	61.0	122	1.0664	0.8125
0.0024	62.0	124	1.0645	0.8125
0.0024	63.0	126	1.0661	0.8125
0.0024	64.0	128	1.0679	0.8125
0.0009	65.0	130	1.0696	0.8125
0.0009	66.0	132	1.0718	0.8125
0.0009	67.0	134	1.0739	0.8125
0.0009	68.0	136	1.0782	0.8125
0.0009	69.0	138	1.0833	0.8125
0.001	70.0	140	1.0910	0.8125
0.001	71.0	142	1.1017	0.8125
0.001	72.0	144	1.1116	0.8125
0.001	73.0	146	1.1187	0.8125
0.001	74.0	148	1.1261	0.8281
0.0019	75.0	150	1.1337	0.8281
0.0019	76.0	152	1.1408	0.8281
0.0019	77.0	154	1.1447	0.8281
0.0019	78.0	156	1.1460	0.8281
0.0019	79.0	158	1.1471	0.8281
0.0007	80.0	160	1.1476	0.8281
0.0007	81.0	162	1.1378	0.8281
0.0007	82.0	164	1.1287	0.8281
0.0007	83.0	166	1.1212	0.8281
0.0007	84.0	168	1.1147	0.8281
0.0007	85.0	170	1.1090	0.8125
0.0007	86.0	172	1.1027	0.8125
0.0007	87.0	174	1.0971	0.8125
0.0007	88.0	176	1.0927	0.8125
0.0007	89.0	178	1.0898	0.8125
0.0006	90.0	180	1.0874	0.8125
0.0006	91.0	182	1.0852	0.8125
0.0006	92.0	184	1.0842	0.8125
0.0006	93.0	186	1.0864	0.8125
0.0006	94.0	188	1.0884	0.8125
0.0006	95.0	190	1.0907	0.8125
0.0006	96.0	192	1.0915	0.8125
0.0006	97.0	194	1.1069	0.8125
0.0006	98.0	196	1.1108	0.8125
0.0006	99.0	198	1.1150	0.8281
0.0025	100.0	200	1.1188	0.8281
0.0025	101.0	202	1.1223	0.8281
0.0025	102.0	204	1.1256	0.8281
0.0025	103.0	206	1.1305	0.8281
0.0025	104.0	208	1.1371	0.8281
0.0005	105.0	210	1.1437	0.8281
0.0005	106.0	212	1.1506	0.8281
0.0005	107.0	214	1.1325	0.8281
0.0005	108.0	216	1.1170	0.8281
0.0005	109.0	218	1.1045	0.8281
0.0004	110.0	220	1.0948	0.8281
0.0004	111.0	222	1.0876	0.8125
0.0004	112.0	224	1.0833	0.8281
0.0004	113.0	226	1.0805	0.8281
0.0004	114.0	228	1.0788	0.8281
0.0004	115.0	230	1.0779	0.8281
0.0004	116.0	232	1.0800	0.8281
0.0004	117.0	234	1.0818	0.8281
0.0004	118.0	236	1.0837	0.8281
0.0004	119.0	238	1.0866	0.8281
0.0004	120.0	240	1.0899	0.8281
0.0004	121.0	242	1.0929	0.8281
0.0004	122.0	244	1.0960	0.8281
0.0004	123.0	246	1.1016	0.8281
0.0004	124.0	248	1.1090	0.8281
0.0003	125.0	250	1.1159	0.8281
0.0003	126.0	252	1.1218	0.8281
0.0003	127.0	254	1.1273	0.8281
0.0003	128.0	256	1.1320	0.8281
0.0003	129.0	258	1.1378	0.8281
0.0003	130.0	260	1.1421	0.8281
0.0003	131.0	262	1.1441	0.8281
0.0003	132.0	264	1.1447	0.8281
0.0003	133.0	266	1.1452	0.8281
0.0003	134.0	268	1.1456	0.8281
0.0003	135.0	270	1.1460	0.8281
0.0003	136.0	272	1.1463	0.8281
0.0003	137.0	274	1.1459	0.8281
0.0003	138.0	276	1.1459	0.8281
0.0003	139.0	278	1.1460	0.8281
0.0003	140.0	280	1.1461	0.8281
0.0003	141.0	282	1.1465	0.8281
0.0003	142.0	284	1.1475	0.8281
0.0003	143.0	286	1.1488	0.8281
0.0003	144.0	288	1.1501	0.8281
0.0002	145.0	290	1.1512	0.8281
0.0002	146.0	292	1.1525	0.8281
0.0002	147.0	294	1.1580	0.8281
0.0002	148.0	296	1.1631	0.8281
0.0002	149.0	298	1.1680	0.8281
0.0002	150.0	300	1.1726	0.8281

Framework versions

• Transformers 4.32.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.4.0
• Tokenizers 0.13.3