logdataset1024samples15epochs_llama3-1_16bit_64eval_adalora_regularized

This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B on the svenwey/LogDataset dataset. It achieves the following results on the evaluation set:

• Loss: 3.1833
• Model Preparation Time: 0.0065
• Linecount Difference Smape Score: 0.8062
• Linecontentlength Difference Smape Score: 0.6578
• Linecontent Sacrebleu Score: 0.5062
• Linecontent Sacrebleu Withoutexplicitnumbers Score: 0.5209
• Timestamps Smape Difference Score: 0.7672
• Timestamps Formatconsistency Score: 1.0
• Timestamps Monotinicity Score: 1.0

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: 5e-05
• train_batch_size: 1
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 4
• 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: linear
• num_epochs: 15.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Linecount Difference Smape Score	Linecontentlength Difference Smape Score	Linecontent Sacrebleu Score	Linecontent Sacrebleu Withoutexplicitnumbers Score	Timestamps Smape Difference Score	Timestamps Formatconsistency Score	Timestamps Monotinicity Score
1.123	0.9996	1157	1.1525	0.0065	0.6605	0.4937	0.3196	0.3222	0.6040	0.9531	0.9531
0.6997	2.0	2315	1.6411	0.0065	0.7375	0.5820	0.4387	0.4382	0.6959	0.9844	0.9844
0.8139	2.9996	3472	1.9592	0.0065	0.7752	0.6222	0.4585	0.4660	0.7382	0.9844	0.9844
0.9399	4.0	4630	2.1386	0.0065	0.8071	0.6818	0.5281	0.5209	0.7880	1.0	1.0
1.0567	4.9996	5787	2.3261	0.0065	0.7964	0.6717	0.5084	0.5090	0.7692	0.9688	0.9688
1.1644	6.0	6945	2.4665	0.0065	0.7903	0.6610	0.5101	0.5111	0.7568	0.9844	0.9844
1.2647	6.9996	8102	2.6250	0.0065	0.7762	0.6409	0.4844	0.5042	0.7473	0.9844	0.9844
1.3484	8.0	9260	2.7645	0.0065	0.7893	0.6484	0.4858	0.5031	0.7561	1.0	1.0
1.4219	8.9996	10417	2.8504	0.0065	0.7926	0.6525	0.4997	0.5011	0.7577	1.0	1.0
1.4816	10.0	11575	2.9482	0.0065	0.7896	0.6619	0.5122	0.5244	0.7632	1.0	0.9844
1.533	10.9996	12732	3.0086	0.0065	0.8268	0.7019	0.5613	0.5720	0.8014	1.0	1.0
1.5711	12.0	13890	3.0883	0.0065	0.8166	0.6826	0.5552	0.5654	0.7954	0.9844	1.0
1.6014	12.9996	15047	3.1250	0.0065	0.8028	0.6716	0.5206	0.5329	0.7780	1.0	1.0
1.619	14.0	16205	3.1550	0.0065	0.7926	0.6482	0.4888	0.4998	0.7604	1.0	1.0
1.6286	14.9935	17355	3.1833	0.0065	0.8062	0.6578	0.5062	0.5209	0.7672	1.0	1.0

Framework versions

• PEFT 0.12.0
• Transformers 4.46.0.dev0
• Pytorch 2.4.1+cu121
• Datasets 2.21.0
• Tokenizers 0.20.0