metadata
license: other
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: deepseek-ai/deepseek-coder-1.3b-base
datasets:
- generator
model-index:
- name: hyperparam-rust-lora-32-3-epoch
results: []
hyperparam-rust-lora-32-3-epoch
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: 0.4542
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: 0.0002
- 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
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.8154 | 0.3 | 25 | 0.5828 |
| 0.5149 | 0.59 | 50 | 0.4992 |
| 0.4731 | 0.89 | 75 | 0.4777 |
| 0.4599 | 1.18 | 100 | 0.4676 |
| 0.443 | 1.48 | 125 | 0.4615 |
| 0.4461 | 1.77 | 150 | 0.4577 |
| 0.4409 | 2.07 | 175 | 0.4553 |
| 0.4375 | 2.37 | 200 | 0.4545 |
| 0.4285 | 2.66 | 225 | 0.4543 |
| 0.4355 | 2.96 | 250 | 0.4542 |
Framework versions
- PEFT 0.10.0
- Transformers 4.39.3
- Pytorch 2.2.1
- Datasets 2.18.0
- Tokenizers 0.15.2