Ip_test_3000

This model is a fine-tuned version of NousResearch/Llama-2-7b-hf on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6617
• Accuracy: 0.6013
• Precision: 0.5938
• Recall: 0.6210
• F1: 0.6071

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: 10
• eval_batch_size: 10
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 160
• 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
• lr_scheduler_warmup_steps: 500
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
No log	0.96	21	1.1577	0.504	0.0	0.0	0.0
No log	1.96	42	0.9227	0.504	0.0	0.0	0.0
No log	2.96	63	0.6936	0.5147	0.5108	0.5081	0.5094
No log	3.96	84	0.6954	0.496	0.4423	0.0618	0.1085
No log	4.96	105	0.6898	0.56	0.5453	0.6801	0.6053
No log	5.96	126	0.6880	0.5653	0.5676	0.5188	0.5421
No log	6.96	147	0.6856	0.5627	0.5780	0.4382	0.4985
13.66	7.96	168	0.6873	0.5573	0.5369	0.7823	0.6368
13.66	8.96	189	0.6793	0.5893	0.5741	0.6667	0.6169
13.66	9.96	210	0.6777	0.584	0.5704	0.6532	0.6090
13.66	10.96	231	0.6690	0.6133	0.5981	0.6720	0.6329
13.66	11.96	252	0.6959	0.5747	0.7087	0.2419	0.3607
13.66	12.96	273	0.6691	0.6093	0.6010	0.6317	0.6160
13.66	13.96	294	0.6689	0.5987	0.5843	0.6613	0.6204
10.9484	14.96	315	0.6635	0.6	0.5984	0.5887	0.5935
10.9484	15.96	336	0.6617	0.6013	0.5938	0.6210	0.6071

Framework versions

• PEFT 0.14.0
• Transformers 4.47.1
• Pytorch 2.3.1.post300
• Datasets 3.2.0
• Tokenizers 0.21.0