jerseyjerry/task-5-microsoft-Phi-3-mini-4k-instruct-20250224

lora

This model is a fine-tuned version of microsoft/Phi-3-mini-4k-instruct on the flock_task5_tranning dataset. It achieves the following results on the evaluation set:

• Loss: 0.7054

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: 2
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 2
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• total_eval_batch_size: 2
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
1.287	0.6897	10	1.4997
1.3264	1.3448	20	1.3904
0.9293	2.0	30	1.2929
1.2359	2.6897	40	1.2001
1.1123	3.3448	50	1.1125
0.8456	4.0	60	1.0130
0.92	4.6897	70	0.9277
0.9875	5.3448	80	0.8583
0.6225	6.0	90	0.8013
0.825	6.6897	100	0.7550
0.6775	7.3448	110	0.7281
0.6614	8.0	120	0.7120
0.6203	8.6897	130	0.7067
0.6521	9.3448	140	0.7054

Framework versions

• PEFT 0.12.0
• Transformers 4.48.3
• Pytorch 2.6.0+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0