metadata
license: apache-2.0
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
datasets:
- code_search_net
base_model: petals-team/falcon-rw-1b
model-index:
- name: falcon-code-generation
results: []
falcon-code-generation
This model is a fine-tuned version of petals-team/falcon-rw-1b on the code_search_net dataset. It achieves the following results on the evaluation set:
- Loss: 0.9946
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: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.03
- training_steps: 320
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.2494 | 0.01 | 20 | 1.2566 |
| 1.2362 | 0.02 | 40 | 1.1664 |
| 1.3029 | 0.03 | 60 | 1.1197 |
| 1.1248 | 0.04 | 80 | 1.0829 |
| 1.2031 | 0.04 | 100 | 1.0871 |
| 1.0559 | 0.05 | 120 | 1.0503 |
| 1.1049 | 0.06 | 140 | 1.0452 |
| 1.1459 | 0.07 | 160 | 1.0415 |
| 1.0521 | 0.08 | 180 | 1.0262 |
| 1.2023 | 0.09 | 200 | 1.0233 |
| 0.9935 | 0.1 | 220 | 1.0112 |
| 1.016 | 0.11 | 240 | 1.0035 |
| 1.1182 | 0.12 | 260 | 0.9993 |
| 1.0044 | 0.12 | 280 | 0.9957 |
| 1.0908 | 0.13 | 300 | 0.9949 |
| 0.9208 | 0.14 | 320 | 0.9946 |
Framework versions
- PEFT 0.10.0
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2