metadata
license: bigcode-openrail-m
base_model: bigcode/starcoder
tags:
- generated_from_trainer
model-index:
- name: starcoder-finetuned-test_newProgram
results: []
starcoder-finetuned-test_newProgram
This model is a fine-tuned version of bigcode/starcoder on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1046
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: 8e-06
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 80
- training_steps: 800
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.1341 | 0.06 | 50 | 0.1215 |
| 0.1238 | 0.12 | 100 | 0.1098 |
| 0.1157 | 1.01 | 150 | 0.1077 |
| 0.1147 | 1.07 | 200 | 0.1068 |
| 0.1147 | 1.13 | 250 | 0.1062 |
| 0.1123 | 2.01 | 300 | 0.1059 |
| 0.1121 | 2.07 | 350 | 0.1055 |
| 0.1126 | 2.14 | 400 | 0.1052 |
| 0.1106 | 3.02 | 450 | 0.1051 |
| 0.1109 | 3.08 | 500 | 0.1049 |
| 0.1125 | 3.14 | 550 | 0.1048 |
| 0.1103 | 4.02 | 600 | 0.1047 |
| 0.1104 | 4.08 | 650 | 0.1047 |
| 0.1118 | 4.15 | 700 | 0.1047 |
| 0.1095 | 5.03 | 750 | 0.1047 |
| 0.1107 | 5.09 | 800 | 0.1046 |
Framework versions
- Transformers 4.32.0.dev0
- Pytorch 2.0.1+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3