best_model-sst-2-64-42

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: 1.4849
• 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	4	1.3914	0.8125
No log	2.0	8	1.3910	0.8203
0.3843	3.0	12	1.3922	0.8203
0.3843	4.0	16	1.3920	0.8203
0.5793	5.0	20	1.3923	0.8203
0.5793	6.0	24	1.3989	0.8203
0.5793	7.0	28	1.4029	0.8281
0.3663	8.0	32	1.4103	0.8281
0.3663	9.0	36	1.3999	0.8281
0.2779	10.0	40	1.4010	0.8281
0.2779	11.0	44	1.3978	0.8281
0.2779	12.0	48	1.3963	0.8203
0.3589	13.0	52	1.4087	0.8203
0.3589	14.0	56	1.4067	0.8281
0.3185	15.0	60	1.4148	0.8281
0.3185	16.0	64	1.4171	0.8359
0.3185	17.0	68	1.4140	0.8359
0.1743	18.0	72	1.3982	0.8359
0.1743	19.0	76	1.3650	0.8359
0.1416	20.0	80	1.3456	0.8359
0.1416	21.0	84	1.3210	0.8359
0.1416	22.0	88	1.3070	0.8359
0.0354	23.0	92	1.3015	0.8359
0.0354	24.0	96	1.3319	0.8438
0.0035	25.0	100	1.3656	0.8281
0.0035	26.0	104	1.3587	0.8281
0.0035	27.0	108	1.3243	0.8359
0.0006	28.0	112	1.2945	0.8438
0.0006	29.0	116	1.2898	0.8438
0.0028	30.0	120	1.3066	0.8438
0.0028	31.0	124	1.3055	0.8438
0.0028	32.0	128	1.3202	0.8438
0.0049	33.0	132	1.3351	0.8438
0.0049	34.0	136	1.3190	0.8438
0.0102	35.0	140	1.3141	0.8438
0.0102	36.0	144	1.3142	0.8438
0.0102	37.0	148	1.3647	0.8281
0.0034	38.0	152	1.4250	0.8203
0.0034	39.0	156	1.4708	0.8203
0.0001	40.0	160	1.4570	0.8203
0.0001	41.0	164	1.4446	0.8203
0.0001	42.0	168	1.4345	0.8281
0.0001	43.0	172	1.4272	0.8281
0.0001	44.0	176	1.4185	0.8281
0.0001	45.0	180	1.4048	0.8281
0.0001	46.0	184	1.3962	0.8281
0.0001	47.0	188	1.4924	0.8203
0.0002	48.0	192	1.5361	0.8125
0.0002	49.0	196	1.5831	0.8125
0.0292	50.0	200	1.4789	0.8281
0.0292	51.0	204	1.2642	0.8359
0.0292	52.0	208	1.2154	0.8516
0.0001	53.0	212	1.1895	0.8516
0.0001	54.0	216	1.1775	0.8438
0.0001	55.0	220	1.1730	0.8438
0.0001	56.0	224	1.1746	0.8438
0.0001	57.0	228	1.1782	0.8516
0.0001	58.0	232	1.1838	0.8516
0.0001	59.0	236	1.2456	0.8281
0.025	60.0	240	1.3887	0.8281
0.025	61.0	244	1.4950	0.8125
0.025	62.0	248	1.5753	0.8047
0.0001	63.0	252	1.6287	0.8047
0.0001	64.0	256	1.6608	0.8047
0.0001	65.0	260	1.6803	0.8047
0.0001	66.0	264	1.6919	0.7969
0.0001	67.0	268	1.5961	0.8047
0.0001	68.0	272	1.4858	0.8125
0.0001	69.0	276	1.4104	0.8281
0.0001	70.0	280	1.3623	0.8281
0.0001	71.0	284	1.3333	0.8359
0.0001	72.0	288	1.3172	0.8359
0.0	73.0	292	1.3107	0.8359
0.0	74.0	296	1.5801	0.8047
0.0014	75.0	300	1.7857	0.8047
0.0014	76.0	304	1.8724	0.7969
0.0014	77.0	308	1.9146	0.7969
0.0001	78.0	312	1.9250	0.7969
0.0001	79.0	316	1.9265	0.7969
0.0001	80.0	320	1.9268	0.7969
0.0001	81.0	324	1.9243	0.7969
0.0001	82.0	328	1.9215	0.7969
0.0	83.0	332	1.9188	0.7969
0.0	84.0	336	1.9159	0.7969
0.0	85.0	340	1.9137	0.7969
0.0	86.0	344	1.9119	0.7969
0.0	87.0	348	1.9103	0.7969
0.0009	88.0	352	1.6541	0.8047
0.0009	89.0	356	1.2749	0.8438
0.0	90.0	360	1.2046	0.8438
0.0	91.0	364	1.1909	0.8438
0.0	92.0	368	1.1860	0.8594
0.0	93.0	372	1.1901	0.8594
0.0	94.0	376	1.1966	0.8516
0.0001	95.0	380	1.2014	0.8516
0.0001	96.0	384	1.2061	0.8438
0.0001	97.0	388	1.2109	0.8438
0.0	98.0	392	1.2170	0.8516
0.0	99.0	396	1.2210	0.8516
0.0	100.0	400	1.2237	0.8516
0.0	101.0	404	1.2258	0.8516
0.0	102.0	408	1.2276	0.8438
0.0	103.0	412	1.2290	0.8438
0.0	104.0	416	1.2301	0.8438
0.0	105.0	420	1.2313	0.8438
0.0	106.0	424	1.2324	0.8438
0.0	107.0	428	1.2334	0.8438
0.0	108.0	432	1.2345	0.8438
0.0	109.0	436	1.2356	0.8438
0.0	110.0	440	1.2366	0.8438
0.0	111.0	444	1.2375	0.8516
0.0	112.0	448	1.2384	0.8516
0.0	113.0	452	1.2400	0.8516
0.0	114.0	456	1.2415	0.8516
0.0	115.0	460	1.2428	0.8516
0.0	116.0	464	1.2439	0.8516
0.0	117.0	468	1.2450	0.8516
0.0	118.0	472	1.2459	0.8516
0.0	119.0	476	1.2467	0.8516
0.0	120.0	480	1.2476	0.8516
0.0	121.0	484	1.2485	0.8516
0.0	122.0	488	1.2495	0.8516
0.0	123.0	492	1.2495	0.8516
0.0	124.0	496	1.2491	0.8516
0.0	125.0	500	1.2491	0.8516
0.0	126.0	504	1.2494	0.8516
0.0	127.0	508	1.2498	0.8516
0.0	128.0	512	1.2503	0.8516
0.0	129.0	516	1.2509	0.8516
0.0	130.0	520	1.2514	0.8516
0.0	131.0	524	1.2519	0.8516
0.0	132.0	528	1.2527	0.8516
0.0	133.0	532	1.2535	0.8516
0.0	134.0	536	1.2542	0.8516
0.0	135.0	540	1.2549	0.8516
0.0	136.0	544	1.2554	0.8516
0.0	137.0	548	1.3879	0.8359
0.0001	138.0	552	1.6893	0.7969
0.0001	139.0	556	1.8348	0.7969
0.0	140.0	560	1.8942	0.7969
0.0	141.0	564	1.8778	0.7969
0.0	142.0	568	1.7187	0.8047
0.0001	143.0	572	1.6119	0.8203
0.0001	144.0	576	1.5523	0.8281
0.0	145.0	580	1.5189	0.8281
0.0	146.0	584	1.5008	0.8281
0.0	147.0	588	1.4916	0.8281
0.0	148.0	592	1.4872	0.8281
0.0	149.0	596	1.4854	0.8281
0.0	150.0	600	1.4849	0.8281

Framework versions

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