smol-135-tq-false

This model is a fine-tuned version of HuggingFaceTB/SmolLM2-135M on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0587
• < Precision: 0.9835
• < Recall: 0.9892
• < F1-score: 0.9863
• < Support: 2588.0

• Precision: 0.9876

• Recall: 0.9846

• F1-score: 0.9861

• Support: 2268.0

• = Precision: 0.8779
• = Recall: 0.8647
• = F1-score: 0.8712
• = Support: 133.0
• • Precision: 0.2
• • Recall: 0.0909
• • F1-score: 0.125
• • Support: 11.0
• Accuracy: 0.9818
• Macro Avg Precision: 0.7622
• Macro Avg Recall: 0.7323
• Macro Avg F1-score: 0.7422
• Macro Avg Support: 5000.0
• Weighted Avg Precision: 0.9808
• Weighted Avg Recall: 0.9818
• Weighted Avg F1-score: 0.9813
• Weighted Avg Support: 5000.0

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: 0.001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 4
• total_train_batch_size: 512
• total_eval_batch_size: 128
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: reduce_lr_on_plateau
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	< Precision	< Recall	< F1-score	< Support	> Precision	> Recall	> F1-score	> Support	= Precision	= Recall	= F1-score	= Support	- Precision	- Recall	- F1-score	- Support	Accuracy	Macro Avg Precision	Macro Avg Recall	Macro Avg F1-score	Macro Avg Support	Weighted Avg Precision	Weighted Avg Recall	Weighted Avg F1-score	Weighted Avg Support
0.4944	1.0	443	0.1333	0.8957	0.9359	0.9153	2588.0	0.9071	0.8955	0.9013	2268.0	0.5965	0.2556	0.3579	133.0	0.0	0.0	0.0	11.0	0.8974	0.5998	0.5217	0.5436	5000.0	0.8909	0.8974	0.8921	5000.0
0.1949	2.0	886	0.0753	0.9411	0.9687	0.9547	2588.0	0.9684	0.9321	0.9499	2268.0	0.5686	0.6541	0.6084	133.0	0.0	0.0	0.0	11.0	0.9416	0.6195	0.6387	0.6282	5000.0	0.9415	0.9416	0.9412	5000.0
0.138	3.0	1329	0.0536	0.9492	0.9826	0.9656	2588.0	0.972	0.9643	0.9681	2268.0	0.8732	0.4662	0.6078	133.0	0.0	0.0	0.0	11.0	0.9584	0.6986	0.6033	0.6354	5000.0	0.9555	0.9584	0.9551	5000.0
0.0787	4.0	1772	0.0360	0.9781	0.9845	0.9813	2588.0	0.9798	0.9824	0.9811	2268.0	0.8291	0.7293	0.776	133.0	0.0	0.0	0.0	11.0	0.9746	0.6967	0.6741	0.6846	5000.0	0.9728	0.9746	0.9736	5000.0
0.0449	5.0	2215	0.0399	0.9826	0.9791	0.9808	2588.0	0.9832	0.9780	0.9805	2268.0	0.7368	0.8421	0.7860	133.0	0.2308	0.2727	0.25	11.0	0.9734	0.7333	0.7680	0.7493	5000.0	0.9746	0.9734	0.9739	5000.0
0.0143	6.0	2658	0.0362	0.9753	0.9915	0.9833	2588.0	0.9888	0.9762	0.9825	2268.0	0.8699	0.8045	0.8359	133.0	0.1429	0.0909	0.1111	11.0	0.9776	0.7442	0.7158	0.7282	5000.0	0.9768	0.9776	0.9771	5000.0
0.008	7.0	3101	0.0400	0.9819	0.9849	0.9834	2588.0	0.9820	0.9850	0.9835	2268.0	0.8468	0.7895	0.8171	133.0	0.0	0.0	0.0	11.0	0.9776	0.7027	0.6899	0.6960	5000.0	0.9762	0.9776	0.9769	5000.0
0.0052	8.0	3544	0.0447	0.9816	0.9880	0.9848	2588.0	0.9846	0.9863	0.9855	2268.0	0.8917	0.8045	0.8458	133.0	0.0	0.0	0.0	11.0	0.9802	0.7145	0.6947	0.7040	5000.0	0.9784	0.9802	0.9792	5000.0
0.0003	9.0	3987	0.0488	0.9816	0.9884	0.9850	2588.0	0.9876	0.9850	0.9863	2268.0	0.8682	0.8421	0.8550	133.0	0.0	0.0	0.0	11.0	0.9808	0.7094	0.7039	0.7066	5000.0	0.9791	0.9808	0.9800	5000.0
0.0002	10.0	4430	0.0500	0.9846	0.9876	0.9861	2588.0	0.9863	0.9859	0.9861	2268.0	0.8769	0.8571	0.8669	133.0	0.1429	0.0909	0.1111	11.0	0.9814	0.7477	0.7304	0.7376	5000.0	0.9807	0.9814	0.9810	5000.0
0.0002	11.0	4873	0.0587	0.9835	0.9892	0.9863	2588.0	0.9876	0.9846	0.9861	2268.0	0.8779	0.8647	0.8712	133.0	0.2	0.0909	0.125	11.0	0.9818	0.7622	0.7323	0.7422	5000.0	0.9808	0.9818	0.9813	5000.0
0.0	12.0	5316	0.0613	0.9835	0.9888	0.9861	2588.0	0.9859	0.9859	0.9859	2268.0	0.896	0.8421	0.8682	133.0	0.2	0.0909	0.125	11.0	0.9816	0.7663	0.7269	0.7413	5000.0	0.9805	0.9816	0.9810	5000.0
0.0	13.0	5759	0.0634	0.9835	0.9884	0.9859	2588.0	0.9868	0.9863	0.9865	2268.0	0.8898	0.8496	0.8692	133.0	0.2	0.0909	0.125	11.0	0.9818	0.7650	0.7288	0.7417	5000.0	0.9807	0.9818	0.9812	5000.0
0.0	14.0	6202	0.0647	0.9835	0.9884	0.9859	2588.0	0.9868	0.9863	0.9865	2268.0	0.8898	0.8496	0.8692	133.0	0.2	0.0909	0.125	11.0	0.9818	0.7650	0.7288	0.7417	5000.0	0.9807	0.9818	0.9812	5000.0
0.0	15.0	6645	0.0663	0.9835	0.9884	0.9859	2588.0	0.9868	0.9863	0.9865	2268.0	0.8898	0.8496	0.8692	133.0	0.2	0.0909	0.125	11.0	0.9818	0.7650	0.7288	0.7417	5000.0	0.9807	0.9818	0.9812	5000.0
0.0	16.0	7088	0.0671	0.9838	0.9880	0.9859	2588.0	0.9868	0.9859	0.9863	2268.0	0.8769	0.8571	0.8669	133.0	0.2	0.0909	0.125	11.0	0.9816	0.7619	0.7305	0.7410	5000.0	0.9806	0.9816	0.9810	5000.0
0.0	17.0	7531	0.0680	0.9838	0.9880	0.9859	2588.0	0.9868	0.9863	0.9865	2268.0	0.8837	0.8571	0.8702	133.0	0.2	0.0909	0.125	11.0	0.9818	0.7636	0.7306	0.7419	5000.0	0.9808	0.9818	0.9812	5000.0
0.0	18.0	7974	0.0689	0.9838	0.9880	0.9859	2588.0	0.9868	0.9859	0.9863	2268.0	0.8769	0.8571	0.8669	133.0	0.2	0.0909	0.125	11.0	0.9816	0.7619	0.7305	0.7410	5000.0	0.9806	0.9816	0.9810	5000.0
0.0029	19.0	8417	0.0696	0.9838	0.9876	0.9857	2588.0	0.9868	0.9859	0.9863	2268.0	0.8702	0.8571	0.8636	133.0	0.2	0.0909	0.125	11.0	0.9814	0.7602	0.7304	0.7402	5000.0	0.9804	0.9814	0.9809	5000.0
0.0	20.0	8860	0.0699	0.9838	0.9876	0.9857	2588.0	0.9868	0.9863	0.9865	2268.0	0.8769	0.8571	0.8669	133.0	0.2	0.0909	0.125	11.0	0.9816	0.7619	0.7305	0.7411	5000.0	0.9806	0.9816	0.9810	5000.0
0.0	21.0	9303	0.0704	0.9835	0.9876	0.9855	2588.0	0.9868	0.9859	0.9863	2268.0	0.8692	0.8496	0.8593	133.0	0.2	0.0909	0.125	11.0	0.9812	0.7599	0.7285	0.7390	5000.0	0.9802	0.9812	0.9806	5000.0
0.0	22.0	9746	0.0707	0.9838	0.9880	0.9859	2588.0	0.9868	0.9859	0.9863	2268.0	0.8769	0.8571	0.8669	133.0	0.2	0.0909	0.125	11.0	0.9816	0.7619	0.7305	0.7410	5000.0	0.9806	0.9816	0.9810	5000.0
0.0	23.0	10189	0.0711	0.9835	0.9876	0.9855	2588.0	0.9868	0.9854	0.9861	2268.0	0.8702	0.8571	0.8636	133.0	0.2	0.0909	0.125	11.0	0.9812	0.7601	0.7303	0.7401	5000.0	0.9802	0.9812	0.9807	5000.0
0.0	24.0	10632	0.0709	0.9838	0.9876	0.9857	2588.0	0.9868	0.9859	0.9863	2268.0	0.8702	0.8571	0.8636	133.0	0.2	0.0909	0.125	11.0	0.9814	0.7602	0.7304	0.7402	5000.0	0.9804	0.9814	0.9809	5000.0
0.0	25.0	11075	0.0715	0.9838	0.9876	0.9857	2588.0	0.9868	0.9859	0.9863	2268.0	0.8702	0.8571	0.8636	133.0	0.2	0.0909	0.125	11.0	0.9814	0.7602	0.7304	0.7402	5000.0	0.9804	0.9814	0.9809	5000.0
0.0	26.0	11518	0.0714	0.9835	0.9876	0.9855	2588.0	0.9868	0.9854	0.9861	2268.0	0.8702	0.8571	0.8636	133.0	0.2	0.0909	0.125	11.0	0.9812	0.7601	0.7303	0.7401	5000.0	0.9802	0.9812	0.9807	5000.0
0.0	27.0	11961	0.0716	0.9838	0.9876	0.9857	2588.0	0.9868	0.9859	0.9863	2268.0	0.8702	0.8571	0.8636	133.0	0.2	0.0909	0.125	11.0	0.9814	0.7602	0.7304	0.7402	5000.0	0.9804	0.9814	0.9809	5000.0
0.0	28.0	12404	0.0715	0.9838	0.9876	0.9857	2588.0	0.9868	0.9859	0.9863	2268.0	0.8702	0.8571	0.8636	133.0	0.2	0.0909	0.125	11.0	0.9814	0.7602	0.7304	0.7402	5000.0	0.9804	0.9814	0.9809	5000.0
0.0	29.0	12847	0.0717	0.9838	0.9880	0.9859	2588.0	0.9868	0.9859	0.9863	2268.0	0.8769	0.8571	0.8669	133.0	0.2	0.0909	0.125	11.0	0.9816	0.7619	0.7305	0.7410	5000.0	0.9806	0.9816	0.9810	5000.0
0.0	29.9328	13260	0.0718	0.9838	0.9876	0.9857	2588.0	0.9868	0.9859	0.9863	2268.0	0.8702	0.8571	0.8636	133.0	0.2	0.0909	0.125	11.0	0.9814	0.7602	0.7304	0.7402	5000.0	0.9804	0.9814	0.9809	5000.0

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 3.0.1
• Tokenizers 0.21.0