|
--- |
|
license: llama2 |
|
base_model: codellama/CodeLlama-7b-hf |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: sql-code-llama-alan |
|
results: [] |
|
library_name: peft |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# sql-code-llama-alan |
|
|
|
This model is a fine-tuned version of [codellama/CodeLlama-7b-hf](https://huggingface.co/codellama/CodeLlama-7b-hf) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.4576 |
|
|
|
## Model description |
|
|
|
More information needed |
|
|
|
## Intended uses & limitations |
|
|
|
More information needed |
|
|
|
## Training and evaluation data |
|
|
|
More information needed |
|
|
|
## Training procedure |
|
|
|
|
|
The following `bitsandbytes` quantization config was used during training: |
|
- quant_method: bitsandbytes |
|
- _load_in_8bit: True |
|
- _load_in_4bit: False |
|
- llm_int8_threshold: 6.0 |
|
- llm_int8_skip_modules: None |
|
- llm_int8_enable_fp32_cpu_offload: False |
|
- llm_int8_has_fp16_weight: False |
|
- bnb_4bit_quant_type: fp4 |
|
- bnb_4bit_use_double_quant: False |
|
- bnb_4bit_compute_dtype: float32 |
|
- bnb_4bit_quant_storage: uint8 |
|
- load_in_4bit: False |
|
- load_in_8bit: True |
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 0.0003 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 128 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 100 |
|
- training_steps: 400 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 2.1992 | 0.0465 | 20 | 2.0335 | |
|
| 1.14 | 0.0931 | 40 | 0.8371 | |
|
| 0.8045 | 0.1396 | 60 | 0.6549 | |
|
| 0.584 | 0.1862 | 80 | 0.5715 | |
|
| 0.3807 | 0.2327 | 100 | 0.5561 | |
|
| 0.5723 | 0.2792 | 120 | 0.5147 | |
|
| 0.4262 | 0.3258 | 140 | 0.5056 | |
|
| 0.6375 | 0.3723 | 160 | 0.5191 | |
|
| 0.4839 | 0.4188 | 180 | 0.4865 | |
|
| 0.3596 | 0.4654 | 200 | 0.4994 | |
|
| 0.5285 | 0.5119 | 220 | 0.4803 | |
|
| 0.4035 | 0.5585 | 240 | 0.4753 | |
|
| 0.6019 | 0.6050 | 260 | 0.4772 | |
|
| 0.4663 | 0.6515 | 280 | 0.4670 | |
|
| 0.345 | 0.6981 | 300 | 0.4746 | |
|
| 0.509 | 0.7446 | 320 | 0.4652 | |
|
| 0.3946 | 0.7912 | 340 | 0.4614 | |
|
| 0.5714 | 0.8377 | 360 | 0.4614 | |
|
| 0.4525 | 0.8842 | 380 | 0.4585 | |
|
| 0.3432 | 0.9308 | 400 | 0.4576 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.6.0.dev0 |
|
- Transformers 4.44.0.dev0 |
|
- Pytorch 2.2.2+cu121 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |
|
|
