my DC sever

https://discord.gg/yaTfFF6Ut2

我正在計畫微調64K指令模型,請幫助我進行計畫

Support me here if you're interested:

Ko-fi: https://ko-fi.com/ogodwin10

Phi-3.5-mini-instruct-24-9-29

This model is a fine-tuned version of microsoft/Phi-3.5-mini-instruct on the lmsys_chat dataset. It achieves the following results on the evaluation set:

  • Loss: 1.3194

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: 0.0001
  • train_batch_size: 2
  • eval_batch_size: 2
  • seed: 42
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 16
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: reduce_lr_on_plateau
  • training_steps: 1000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
1.428 0.0160 100 1.3194
1.2653 0.0320 200 1.3194
1.3084 0.0480 300 1.3194
1.3234 0.0641 400 1.3194
1.4091 0.0801 500 1.3194
1.2878 0.0961 600 1.3194
1.2933 0.1121 700 1.3194
1.3246 0.1281 800 1.3194
1.2911 0.1441 900 1.3194
1.4227 0.1602 1000 1.3194

Framework versions

  • Transformers 4.45.0
  • Pytorch 2.4.0+cu124
  • Datasets 2.19.1
  • Tokenizers 0.20.0
Downloads last month
24
Safetensors
Model size
3.82B params
Tensor type
F32
·
FP16
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for win10/Phi-3.5-mini-instruct-24-9-29

Finetuned
(50)
this model
Merges
4 models