Modelpublic

train-train/business-news-generator

sourceHugging Faceapache-2.0updated 5mo agoView on Hugging Face

0likes16downloads

Model Card

business-news-generator

This model is a fine-tuned version of HuggingFaceTB/SmolLM-135M on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 5.0033

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: 0.0005
—trainbatchsize: 8
—evalbatchsize: 8
—seed: 42
—optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
—lrschedulertype: cosine
—num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
3.1489	0.32	200	3.2645
2.859	0.64	400	3.2011
2.7303	0.96	600	3.1581
1.8257	1.28	800	3.3220
1.7536	1.6	1000	3.3402
1.7524	1.92	1200	3.3294
1.1166	2.24	1400	3.8387
0.9383	2.56	1600	3.8604
0.9853	2.88	1800	3.8619
0.6414	3.2	2000	4.3422
0.4725	3.52	2200	4.3974
0.5086	3.84	2400	4.3799
0.3855	4.16	2600	4.6208
0.2775	4.48	2800	4.6318
0.2811	4.8	3000	4.6108
0.2326	5.12	3200	4.7385
0.1644	5.44	3400	4.7339
0.1742	5.76	3600	4.7361
0.1585	6.08	3800	4.8381
0.1076	6.4	4000	4.8383
0.1149	6.72	4200	4.8113
0.1122	7.04	4400	4.8283
0.0821	7.36	4600	4.8999
0.0822	7.68	4800	4.9060
0.0863	8.0	5000	4.9062
0.0655	8.32	5200	4.9783
0.0703	8.64	5400	4.9760
0.071	8.96	5600	4.9732
0.0613	9.28	5800	4.9946
0.0606	9.6	6000	5.0015
0.061	9.92	6200	5.0033

Framework versions

—Transformers 4.57.6
—Pytorch 2.10.0+cu128
—Datasets 4.0.0
—Tokenizers 0.22.2