best_model-sst-2-32-100

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: 0.5168
• Accuracy: 0.9219

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	0.8101	0.9062
No log	2.0	4	0.8102	0.9062
No log	3.0	6	0.8102	0.9062
No log	4.0	8	0.8100	0.9062
0.6019	5.0	10	0.8098	0.9062
0.6019	6.0	12	0.8095	0.9062
0.6019	7.0	14	0.8090	0.9062
0.6019	8.0	16	0.8085	0.9062
0.6019	9.0	18	0.8079	0.9062
0.6181	10.0	20	0.8073	0.9062
0.6181	11.0	22	0.8066	0.9062
0.6181	12.0	24	0.8061	0.9062
0.6181	13.0	26	0.8055	0.9062
0.6181	14.0	28	0.8048	0.9062
0.5045	15.0	30	0.8037	0.9062
0.5045	16.0	32	0.8020	0.9062
0.5045	17.0	34	0.8003	0.9062
0.5045	18.0	36	0.7978	0.9062
0.5045	19.0	38	0.7955	0.9062
0.4784	20.0	40	0.7928	0.9062
0.4784	21.0	42	0.7902	0.9062
0.4784	22.0	44	0.7868	0.9062
0.4784	23.0	46	0.7824	0.9062
0.4784	24.0	48	0.7764	0.9062
0.3582	25.0	50	0.7695	0.9062
0.3582	26.0	52	0.7628	0.9062
0.3582	27.0	54	0.7548	0.9062
0.3582	28.0	56	0.7473	0.9062
0.3582	29.0	58	0.7388	0.9062
0.3152	30.0	60	0.7286	0.9062
0.3152	31.0	62	0.7145	0.9062
0.3152	32.0	64	0.7007	0.9062
0.3152	33.0	66	0.6860	0.9062
0.3152	34.0	68	0.6662	0.9062
0.2403	35.0	70	0.6377	0.9062
0.2403	36.0	72	0.5941	0.9062
0.2403	37.0	74	0.5458	0.8906
0.2403	38.0	76	0.4985	0.8906
0.2403	39.0	78	0.4676	0.9219
0.1021	40.0	80	0.4598	0.9219
0.1021	41.0	82	0.4572	0.9375
0.1021	42.0	84	0.4521	0.9375
0.1021	43.0	86	0.4493	0.9375
0.1021	44.0	88	0.4420	0.9375
0.016	45.0	90	0.4264	0.9375
0.016	46.0	92	0.4104	0.9375
0.016	47.0	94	0.4008	0.9375
0.016	48.0	96	0.4056	0.9062
0.016	49.0	98	0.4256	0.9219
0.0016	50.0	100	0.4450	0.9062
0.0016	51.0	102	0.4667	0.9062
0.0016	52.0	104	0.4946	0.9062
0.0016	53.0	106	0.5189	0.9062
0.0016	54.0	108	0.5347	0.9062
0.0008	55.0	110	0.5434	0.9062
0.0008	56.0	112	0.5500	0.9062
0.0008	57.0	114	0.5545	0.9062
0.0008	58.0	116	0.5557	0.9062
0.0008	59.0	118	0.5535	0.9062
0.0005	60.0	120	0.5492	0.9062
0.0005	61.0	122	0.5389	0.9062
0.0005	62.0	124	0.5249	0.9062
0.0005	63.0	126	0.5044	0.9062
0.0005	64.0	128	0.4804	0.9062
0.0008	65.0	130	0.4611	0.9219
0.0008	66.0	132	0.4474	0.9375
0.0008	67.0	134	0.4373	0.9375
0.0008	68.0	136	0.4299	0.9375
0.0008	69.0	138	0.4246	0.9219
0.0003	70.0	140	0.4213	0.9219
0.0003	71.0	142	0.4191	0.9219
0.0003	72.0	144	0.4177	0.9219
0.0003	73.0	146	0.4283	0.9219
0.0003	74.0	148	0.4393	0.9375
0.0011	75.0	150	0.4489	0.9375
0.0011	76.0	152	0.4577	0.9375
0.0011	77.0	154	0.4659	0.9375
0.0011	78.0	156	0.4734	0.9219
0.0011	79.0	158	0.4803	0.9219
0.0003	80.0	160	0.4866	0.9219
0.0003	81.0	162	0.4924	0.9062
0.0003	82.0	164	0.4845	0.9219
0.0003	83.0	166	0.4663	0.9375
0.0003	84.0	168	0.4532	0.9375
0.0072	85.0	170	0.4429	0.9375
0.0072	86.0	172	0.4352	0.9375
0.0072	87.0	174	0.4297	0.9375
0.0072	88.0	176	0.4255	0.9219
0.0072	89.0	178	0.4223	0.9219
0.0002	90.0	180	0.4201	0.9219
0.0002	91.0	182	0.4184	0.9219
0.0002	92.0	184	0.4171	0.9219
0.0002	93.0	186	0.4163	0.9219
0.0002	94.0	188	0.4231	0.9219
0.0002	95.0	190	0.4306	0.9375
0.0002	96.0	192	0.4377	0.9375
0.0002	97.0	194	0.4440	0.9375
0.0002	98.0	196	0.4494	0.9375
0.0002	99.0	198	0.4542	0.9375
0.0002	100.0	200	0.4582	0.9375
0.0002	101.0	202	0.4617	0.9375
0.0002	102.0	204	0.4646	0.9375
0.0002	103.0	206	0.4676	0.9375
0.0002	104.0	208	0.4705	0.9375
0.0002	105.0	210	0.4729	0.9375
0.0002	106.0	212	0.4749	0.9375
0.0002	107.0	214	0.4769	0.9375
0.0002	108.0	216	0.4788	0.9375
0.0002	109.0	218	0.4803	0.9375
0.0002	110.0	220	0.4810	0.9375
0.0002	111.0	222	0.4817	0.9375
0.0002	112.0	224	0.4825	0.9375
0.0002	113.0	226	0.4837	0.9375
0.0002	114.0	228	0.4849	0.9375
0.0002	115.0	230	0.4857	0.9219
0.0002	116.0	232	0.4679	0.9375
0.0002	117.0	234	0.4374	0.9375
0.0002	118.0	236	0.4225	0.9375
0.0002	119.0	238	0.4275	0.9375
0.0004	120.0	240	0.4352	0.9375
0.0004	121.0	242	0.4423	0.9375
0.0004	122.0	244	0.4481	0.9375
0.0004	123.0	246	0.4509	0.9375
0.0004	124.0	248	0.4527	0.9375
0.0002	125.0	250	0.4528	0.9375
0.0002	126.0	252	0.4530	0.9375
0.0002	127.0	254	0.4531	0.9375
0.0002	128.0	256	0.4531	0.9375
0.0002	129.0	258	0.4530	0.9375
0.0014	130.0	260	0.4188	0.9375
0.0014	131.0	262	0.4099	0.9531
0.0014	132.0	264	0.4306	0.9219
0.0014	133.0	266	0.4583	0.9219
0.0014	134.0	268	0.4801	0.9219
0.0001	135.0	270	0.4951	0.9219
0.0001	136.0	272	0.5056	0.9219
0.0001	137.0	274	0.5134	0.9062
0.0001	138.0	276	0.5179	0.9062
0.0001	139.0	278	0.5215	0.9062
0.0001	140.0	280	0.5243	0.9062
0.0001	141.0	282	0.5255	0.9062
0.0001	142.0	284	0.5258	0.9062
0.0001	143.0	286	0.5261	0.9062
0.0001	144.0	288	0.5262	0.9062
0.0001	145.0	290	0.5261	0.9062
0.0001	146.0	292	0.5236	0.9062
0.0001	147.0	294	0.5214	0.9062
0.0001	148.0	296	0.5194	0.9219
0.0001	149.0	298	0.5177	0.9219
0.0001	150.0	300	0.5168	0.9219

Framework versions

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