outputs

This model is a fine-tuned version of microsoft/Phi-3.5-mini-instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.6456

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: 0.0009
• train_batch_size: 4
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 15
• training_steps: 150

Training results

Training Loss	Epoch	Step	Validation Loss
1.8177	1.1111	5	1.4564
1.1218	2.2222	10	0.9293
0.8806	3.3333	15	0.8302
0.6797	4.4444	20	0.8546
0.4134	5.5556	25	0.9876
0.1811	6.6667	30	1.2165
0.162	7.7778	35	1.3668
0.11	8.8889	40	1.5960
0.0843	10.0	45	1.4322
0.0495	11.1111	50	1.4248
0.0338	12.2222	55	1.4805
0.024	13.3333	60	1.6548
0.0365	14.4444	65	1.6456

Framework versions

• PEFT 0.12.0
• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1