ST1_modernbert-base_hazard-category_V1

This model is a fine-tuned version of answerdotai/ModernBERT-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7555
• F1: 0.9462

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: 5e-05
• train_batch_size: 36
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 200

Training results

Training Loss	Epoch	Step	Validation Loss	F1
0.83	1.0	128	0.2556	0.9226
0.2639	2.0	256	0.2905	0.9053
0.17	3.0	384	0.2935	0.9388
0.0967	4.0	512	0.3408	0.9291
0.0549	5.0	640	0.3093	0.9442
0.0523	6.0	768	0.3185	0.9470
0.0312	7.0	896	0.4171	0.9407
0.0165	8.0	1024	0.4558	0.9469
0.0116	9.0	1152	0.4671	0.9359
0.0077	10.0	1280	0.4365	0.9511
0.0071	11.0	1408	0.5180	0.9387
0.008	12.0	1536	0.5229	0.9383
0.001	13.0	1664	0.5507	0.9378
0.0021	14.0	1792	0.5437	0.9427
0.0009	15.0	1920	0.5607	0.9466
0.0001	16.0	2048	0.5666	0.9445
0.0006	17.0	2176	0.5684	0.9405
0.0008	18.0	2304	0.5755	0.9446
0.0004	19.0	2432	0.5756	0.9385
0.0004	20.0	2560	0.5775	0.9425
0.0007	21.0	2688	0.5859	0.9389
0.0007	22.0	2816	0.5852	0.9385
0.0007	23.0	2944	0.5931	0.9427
0.0003	24.0	3072	0.5948	0.9408
0.0007	25.0	3200	0.5976	0.9408
0.0006	26.0	3328	0.5907	0.9385
0.0	27.0	3456	0.5985	0.9445
0.0009	28.0	3584	0.6052	0.9409
0.0005	29.0	3712	0.5952	0.9464
0.0006	30.0	3840	0.6037	0.9445
0.0	31.0	3968	0.5994	0.9445
0.0006	32.0	4096	0.6061	0.9409
0.0005	33.0	4224	0.6046	0.9445
0.0006	34.0	4352	0.6007	0.9445
0.0	35.0	4480	0.6052	0.9445
0.0006	36.0	4608	0.6091	0.9425
0.0003	37.0	4736	0.6106	0.9445
0.0008	38.0	4864	0.6117	0.9445
0.0002	39.0	4992	0.6054	0.9464
0.0	40.0	5120	0.6080	0.9443
0.0004	41.0	5248	0.6117	0.9443
0.0011	42.0	5376	0.6152	0.9443
0.0004	43.0	5504	0.6164	0.9425
0.0007	44.0	5632	0.6188	0.9425
0.0005	45.0	5760	0.6104	0.9442
0.0	46.0	5888	0.6164	0.9446
0.0006	47.0	6016	0.6104	0.9439
0.0002	48.0	6144	0.6129	0.9443
0.0007	49.0	6272	0.6205	0.9424
0.0006	50.0	6400	0.6182	0.9425
0.0006	51.0	6528	0.6113	0.9439
0.0	52.0	6656	0.6212	0.9443
0.0005	53.0	6784	0.6186	0.9440
0.0004	54.0	6912	0.6163	0.9440
0.0	55.0	7040	0.6172	0.9440
0.0008	56.0	7168	0.6163	0.9440
0.0005	57.0	7296	0.6211	0.9440
0.0006	58.0	7424	0.6232	0.9422
0.0003	59.0	7552	0.6240	0.9440
0.0003	60.0	7680	0.6224	0.9440
0.0005	61.0	7808	0.6273	0.9419
0.0006	62.0	7936	0.6239	0.9423
0.0001	63.0	8064	0.6216	0.9419
0.0004	64.0	8192	0.6191	0.9420
0.0005	65.0	8320	0.6169	0.9420
0.0	66.0	8448	0.6201	0.9420
0.0005	67.0	8576	0.6218	0.9402
0.0004	68.0	8704	0.6195	0.9421
0.0004	69.0	8832	0.6246	0.9402
0.0002	70.0	8960	0.6269	0.9420
0.0003	71.0	9088	0.6268	0.9402
0.0005	72.0	9216	0.6254	0.9418
0.0	73.0	9344	0.6273	0.9402
0.0007	74.0	9472	0.6257	0.9437
0.0005	75.0	9600	0.6213	0.9399
0.0005	76.0	9728	0.6266	0.9418
0.0002	77.0	9856	0.6258	0.9418
0.0005	78.0	9984	0.6298	0.9418
0.0003	79.0	10112	0.6242	0.9439
0.0002	80.0	10240	0.6284	0.9418
0.0008	81.0	10368	0.6255	0.9439
0.0	82.0	10496	0.6312	0.9439
0.0005	83.0	10624	0.6312	0.9399
0.0002	84.0	10752	0.6279	0.9381
0.0005	85.0	10880	0.6295	0.9401
0.0005	86.0	11008	0.6231	0.9433
0.0005	87.0	11136	0.6302	0.9433
0.0002	88.0	11264	0.6281	0.9433
0.0003	89.0	11392	0.6326	0.9433
0.0002	90.0	11520	0.6347	0.9418
0.0005	91.0	11648	0.6324	0.9418
0.0007	92.0	11776	0.6362	0.9418
0.0005	93.0	11904	0.6351	0.9433
0.0004	94.0	12032	0.6372	0.9433
0.0002	95.0	12160	0.6347	0.9433
0.0005	96.0	12288	0.6378	0.9418
0.0005	97.0	12416	0.6384	0.9418
0.0005	98.0	12544	0.6449	0.9418
0.0	99.0	12672	0.6418	0.9433
0.0005	100.0	12800	0.6540	0.9454
0.0005	101.0	12928	0.6413	0.9466
0.0111	102.0	13056	0.5095	0.9187
0.0623	103.0	13184	0.5184	0.9350
0.0167	104.0	13312	0.5990	0.9222
0.0052	105.0	13440	0.6861	0.9409
0.0066	106.0	13568	0.6613	0.9455
0.0003	107.0	13696	0.6736	0.9462
0.0002	108.0	13824	0.6888	0.9446
0.0005	109.0	13952	0.6931	0.9462
0.0004	110.0	14080	0.6953	0.9462
0.0002	111.0	14208	0.6987	0.9462
0.0002	112.0	14336	0.7009	0.9462
0.0006	113.0	14464	0.7038	0.9462
0.0	114.0	14592	0.7079	0.9462
0.0004	115.0	14720	0.7073	0.9462
0.0	116.0	14848	0.7094	0.9462
0.0008	117.0	14976	0.7091	0.9462
0.0004	118.0	15104	0.7108	0.9462
0.0004	119.0	15232	0.7111	0.9462
0.0002	120.0	15360	0.7138	0.9462
0.0004	121.0	15488	0.7149	0.9462
0.0004	122.0	15616	0.7144	0.9462
0.0004	123.0	15744	0.7164	0.9462
0.0004	124.0	15872	0.7178	0.9462
0.0004	125.0	16000	0.7178	0.9462
0.0002	126.0	16128	0.7191	0.9462
0.0007	127.0	16256	0.7189	0.9462
0.0002	128.0	16384	0.7203	0.9462
0.0004	129.0	16512	0.7215	0.9462
0.0	130.0	16640	0.7221	0.9462
0.0009	131.0	16768	0.7232	0.9462
0.0	132.0	16896	0.7236	0.9462
0.0002	133.0	17024	0.7242	0.9462
0.0002	134.0	17152	0.7253	0.9462
0.0004	135.0	17280	0.7247	0.9462
0.0002	136.0	17408	0.7248	0.9462
0.0004	137.0	17536	0.7265	0.9462
0.0002	138.0	17664	0.7264	0.9462
0.0003	139.0	17792	0.7306	0.9462
0.0004	140.0	17920	0.7302	0.9462
0.0002	141.0	18048	0.7304	0.9462
0.0004	142.0	18176	0.7307	0.9462
0.0002	143.0	18304	0.7325	0.9462
0.0002	144.0	18432	0.7324	0.9462
0.0007	145.0	18560	0.7321	0.9462
0.0	146.0	18688	0.7324	0.9462
0.0004	147.0	18816	0.7355	0.9462
0.0004	148.0	18944	0.7348	0.9462
0.0002	149.0	19072	0.7355	0.9462
0.0004	150.0	19200	0.7357	0.9462
0.0004	151.0	19328	0.7371	0.9462
0.0002	152.0	19456	0.7374	0.9462
0.0004	153.0	19584	0.7384	0.9462
0.0002	154.0	19712	0.7387	0.9462
0.0004	155.0	19840	0.7390	0.9462
0.0004	156.0	19968	0.7396	0.9462
0.0002	157.0	20096	0.7400	0.9462
0.0002	158.0	20224	0.7420	0.9462
0.0004	159.0	20352	0.7391	0.9462
0.0006	160.0	20480	0.7420	0.9462
0.0004	161.0	20608	0.7428	0.9462
0.0004	162.0	20736	0.7436	0.9462
0.0002	163.0	20864	0.7442	0.9462
0.0002	164.0	20992	0.7444	0.9462
0.0004	165.0	21120	0.7451	0.9446
0.0002	166.0	21248	0.7450	0.9462
0.0004	167.0	21376	0.7452	0.9462
0.0002	168.0	21504	0.7478	0.9462
0.0004	169.0	21632	0.7467	0.9462
0.0002	170.0	21760	0.7467	0.9480
0.0006	171.0	21888	0.7491	0.9462
0.0004	172.0	22016	0.7489	0.9462
0.0002	173.0	22144	0.7491	0.9462
0.0004	174.0	22272	0.7503	0.9462
0.0002	175.0	22400	0.7513	0.9462
0.0004	176.0	22528	0.7496	0.9462
0.0	177.0	22656	0.7511	0.9462
0.0006	178.0	22784	0.7508	0.9480
0.0002	179.0	22912	0.7519	0.9462
0.0004	180.0	23040	0.7535	0.9462
0.0002	181.0	23168	0.7534	0.9462
0.0002	182.0	23296	0.7530	0.9462
0.0004	183.0	23424	0.7522	0.9480
0.0002	184.0	23552	0.7524	0.9462
0.0002	185.0	23680	0.7529	0.9462
0.0	186.0	23808	0.7536	0.9462
0.0002	187.0	23936	0.7550	0.9480
0.0006	188.0	24064	0.7549	0.9462
0.0	189.0	24192	0.7532	0.9480
0.0004	190.0	24320	0.7556	0.9462
0.0004	191.0	24448	0.7546	0.9462
0.0002	192.0	24576	0.7553	0.9462
0.0004	193.0	24704	0.7571	0.9462
0.0002	194.0	24832	0.7551	0.9480
0.0006	195.0	24960	0.7559	0.9480
0.0002	196.0	25088	0.7552	0.9462
0.0002	197.0	25216	0.7560	0.9480
0.0002	198.0	25344	0.7562	0.9480
0.0004	199.0	25472	0.7552	0.9480
0.0002	200.0	25600	0.7555	0.9462

Framework versions

• Transformers 4.48.0.dev0
• Pytorch 2.4.1+cu121
• Datasets 3.1.0
• Tokenizers 0.21.0