Edit model card

hyperparam-rust-sft-lora

This model is a fine-tuned version of deepseek-ai/deepseek-coder-1.3b-base on the generator dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4247

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: 0.0003
  • train_batch_size: 1
  • eval_batch_size: 1
  • seed: 42
  • gradient_accumulation_steps: 32
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.05
  • lr_scheduler_warmup_steps: 20
  • num_epochs: 5

Training results

Training Loss Epoch Step Validation Loss
0.7919 0.3 25 0.5285
0.4811 0.59 50 0.4738
0.4512 0.89 75 0.4567
0.4367 1.18 100 0.4465
0.4162 1.48 125 0.4399
0.4188 1.77 150 0.4352
0.4127 2.07 175 0.4318
0.3981 2.37 200 0.4296
0.3887 2.66 225 0.4281
0.3943 2.96 250 0.4258
0.3808 3.25 275 0.4263
0.3836 3.55 300 0.4251
0.3824 3.84 325 0.4247
0.3782 4.14 350 0.4246
0.377 4.43 375 0.4247
0.3725 4.73 400 0.4247

Framework versions

  • PEFT 0.10.0
  • Transformers 4.39.3
  • Pytorch 2.2.1
  • Datasets 2.18.0
  • Tokenizers 0.15.2
Downloads last month
14
Inference API
Unable to determine this model’s pipeline type. Check the docs .

Model tree for ysr/hyperparam-rust-sft-lora

Adapter
(100)
this model