best_model-sst-2-64-87

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: 1.2746
• Accuracy: 0.8438

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	1.3247	0.8438
No log	2.0	8	1.3227	0.8438
0.7148	3.0	12	1.3195	0.8438
0.7148	4.0	16	1.3169	0.8359
0.6114	5.0	20	1.3149	0.8359
0.6114	6.0	24	1.3101	0.8359
0.6114	7.0	28	1.2982	0.8438
0.5794	8.0	32	1.2836	0.8438
0.5794	9.0	36	1.2655	0.8438
0.5231	10.0	40	1.2497	0.8438
0.5231	11.0	44	1.2410	0.8438
0.5231	12.0	48	1.2307	0.8438
0.4052	13.0	52	1.2154	0.8438
0.4052	14.0	56	1.2001	0.8438
0.363	15.0	60	1.1877	0.8438
0.363	16.0	64	1.1760	0.8516
0.363	17.0	68	1.1836	0.8516
0.2969	18.0	72	1.1848	0.8594
0.2969	19.0	76	1.1823	0.8516
0.1866	20.0	80	1.1867	0.8516
0.1866	21.0	84	1.1795	0.8516
0.1866	22.0	88	1.1756	0.8516
0.1502	23.0	92	1.1731	0.8516
0.1502	24.0	96	1.1680	0.8516
0.0974	25.0	100	1.1489	0.8516
0.0974	26.0	104	1.1088	0.8516
0.0974	27.0	108	1.0986	0.8594
0.0992	28.0	112	1.0879	0.8594
0.0992	29.0	116	1.0850	0.8594
0.0065	30.0	120	1.1056	0.8594
0.0065	31.0	124	1.1355	0.8516
0.0065	32.0	128	1.1457	0.8438
0.0185	33.0	132	1.1518	0.8438
0.0185	34.0	136	1.1437	0.8438
0.0123	35.0	140	1.1230	0.8516
0.0123	36.0	144	1.1109	0.8516
0.0123	37.0	148	1.1093	0.8594
0.0001	38.0	152	1.1085	0.8594
0.0001	39.0	156	1.1092	0.8594
0.008	40.0	160	1.1163	0.8594
0.008	41.0	164	1.1272	0.8516
0.008	42.0	168	1.1351	0.8516
0.0001	43.0	172	1.1365	0.8516
0.0001	44.0	176	1.1287	0.8516
0.0007	45.0	180	1.1195	0.8594
0.0007	46.0	184	1.1110	0.8594
0.0007	47.0	188	1.1261	0.8594
0.0003	48.0	192	1.1236	0.8594
0.0003	49.0	196	1.1083	0.8594
0.0018	50.0	200	1.1057	0.8594
0.0018	51.0	204	1.1077	0.8594
0.0018	52.0	208	1.1095	0.8516
0.0001	53.0	212	1.1116	0.8594
0.0001	54.0	216	1.1149	0.8594
0.0017	55.0	220	1.1500	0.8516
0.0017	56.0	224	1.1396	0.8516
0.0017	57.0	228	1.1474	0.8516
0.0002	58.0	232	1.1402	0.8594
0.0002	59.0	236	1.1367	0.8594
0.0001	60.0	240	1.1349	0.8516
0.0001	61.0	244	1.1350	0.8516
0.0001	62.0	248	1.1366	0.8516
0.0001	63.0	252	1.1389	0.8594
0.0001	64.0	256	1.1395	0.8594
0.0001	65.0	260	1.1380	0.8594
0.0001	66.0	264	1.1378	0.8594
0.0001	67.0	268	1.1411	0.8594
0.0001	68.0	272	1.1439	0.8594
0.0001	69.0	276	1.1452	0.8594
0.0122	70.0	280	1.1270	0.8594
0.0122	71.0	284	1.1514	0.8594
0.0122	72.0	288	1.1908	0.8516
0.0001	73.0	292	1.2155	0.8516
0.0001	74.0	296	1.2281	0.8516
0.0001	75.0	300	1.2353	0.8516
0.0001	76.0	304	1.2387	0.8516
0.0001	77.0	308	1.2380	0.8516
0.0177	78.0	312	1.1050	0.8594
0.0177	79.0	316	1.1201	0.8594
0.0123	80.0	320	1.1227	0.8516
0.0123	81.0	324	1.1249	0.8594
0.0123	82.0	328	1.1305	0.8594
0.0001	83.0	332	1.1371	0.8672
0.0001	84.0	336	1.1424	0.8672
0.0001	85.0	340	1.1449	0.8672
0.0001	86.0	344	1.1464	0.8672
0.0001	87.0	348	1.1469	0.8672
0.0001	88.0	352	1.1448	0.8594
0.0001	89.0	356	1.1444	0.8594
0.0	90.0	360	1.1452	0.8594
0.0	91.0	364	1.1464	0.8594
0.0	92.0	368	1.1484	0.8594
0.0001	93.0	372	1.1504	0.8594
0.0001	94.0	376	1.1521	0.8516
0.0	95.0	380	1.1537	0.8516
0.0	96.0	384	1.1553	0.8516
0.0	97.0	388	1.1571	0.8516
0.0001	98.0	392	1.1605	0.8594
0.0001	99.0	396	1.1645	0.8594
0.0	100.0	400	1.1678	0.8594
0.0	101.0	404	1.1706	0.8594
0.0	102.0	408	1.1729	0.8594
0.0	103.0	412	1.1747	0.8594
0.0	104.0	416	1.1762	0.8594
0.0001	105.0	420	1.1777	0.8594
0.0001	106.0	424	1.1792	0.8594
0.0001	107.0	428	1.1808	0.8594
0.0034	108.0	432	1.2561	0.8516
0.0034	109.0	436	1.3098	0.8516
0.0063	110.0	440	1.2197	0.8516
0.0063	111.0	444	1.1982	0.8516
0.0063	112.0	448	1.2230	0.8516
0.0	113.0	452	1.2172	0.8594
0.0	114.0	456	1.2165	0.8516
0.0	115.0	460	1.2187	0.8516
0.0	116.0	464	1.2213	0.8516
0.0	117.0	468	1.2234	0.8516
0.0	118.0	472	1.2248	0.8516
0.0	119.0	476	1.2267	0.8516
0.0	120.0	480	1.2288	0.8594
0.0	121.0	484	1.2316	0.8594
0.0	122.0	488	1.2342	0.8594
0.0	123.0	492	1.2364	0.8594
0.0	124.0	496	1.2436	0.8594
0.001	125.0	500	1.2770	0.8438
0.001	126.0	504	1.3138	0.8594
0.001	127.0	508	1.3084	0.8594
0.0	128.0	512	1.3102	0.8438
0.0	129.0	516	1.3333	0.8438
0.0002	130.0	520	1.3251	0.8516
0.0002	131.0	524	1.2928	0.8594
0.0002	132.0	528	1.2468	0.8438
0.0	133.0	532	1.2295	0.8438
0.0	134.0	536	1.2483	0.8438
0.0	135.0	540	1.2652	0.8438
0.0	136.0	544	1.2741	0.8438
0.0	137.0	548	1.2786	0.8438
0.0	138.0	552	1.2811	0.8438
0.0	139.0	556	1.2824	0.8438
0.0	140.0	560	1.2833	0.8438
0.0	141.0	564	1.2837	0.8438
0.0	142.0	568	1.2833	0.8438
0.0	143.0	572	1.2830	0.8438
0.0	144.0	576	1.2828	0.8438
0.0	145.0	580	1.2827	0.8438
0.0	146.0	584	1.2827	0.8438
0.0	147.0	588	1.2827	0.8438
0.0001	148.0	592	1.2786	0.8438
0.0001	149.0	596	1.2755	0.8438
0.0	150.0	600	1.2746	0.8438

Framework versions

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