|
--- |
|
license: llama2 |
|
library_name: peft |
|
tags: |
|
- axolotl |
|
- generated_from_trainer |
|
base_model: codellama/CodeLlama-7b-hf |
|
model-index: |
|
- name: EvolCodeLlama-7b |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl) |
|
<details><summary>See axolotl config</summary> |
|
|
|
axolotl version: `0.4.0` |
|
```yaml |
|
base_model: codellama/CodeLlama-7b-hf |
|
base_model_config: codellama/CodeLlama-7b-hf |
|
model_type: LlamaForCausalLM |
|
tokenizer_type: LlamaTokenizer |
|
is_llama_derived_model: true |
|
hub_model_id: EvolCodeLlama-7b |
|
|
|
load_in_8bit: false |
|
load_in_4bit: true |
|
strict: false |
|
|
|
datasets: |
|
- path: ptoro/Evol-Instruct-Python-1k-testing |
|
type: alpaca |
|
dataset_prepared_path: last_run_prepared |
|
val_set_size: 0.02 |
|
output_dir: ./qlora-out |
|
|
|
adapter: qlora |
|
lora_model_dir: |
|
|
|
sequence_len: 2048 |
|
sample_packing: true |
|
|
|
lora_r: 32 |
|
lora_alpha: 16 |
|
lora_dropout: 0.05 |
|
lora_target_modules: |
|
lora_target_linear: true |
|
lora_fan_in_fan_out: |
|
|
|
wandb_project: axolotl |
|
wandb_entity: |
|
wandb_watch: |
|
wandb_run_id: |
|
wandb_log_model: |
|
|
|
gradient_accumulation_steps: 4 |
|
micro_batch_size: 2 |
|
num_epochs: 3 |
|
optimizer: paged_adamw_32bit |
|
lr_scheduler: cosine |
|
learning_rate: 0.0002 |
|
|
|
train_on_inputs: false |
|
group_by_length: false |
|
bf16: true |
|
fp16: false |
|
tf32: false |
|
|
|
gradient_checkpointing: true |
|
early_stopping_patience: |
|
resume_from_checkpoint: |
|
local_rank: |
|
logging_steps: 1 |
|
xformers_attention: |
|
flash_attention: true |
|
|
|
warmup_steps: 100 |
|
eval_steps: 0.01 |
|
save_strategy: epoch |
|
save_steps: |
|
debug: |
|
deepspeed: |
|
weight_decay: 0.0 |
|
fsdp: |
|
fsdp_config: |
|
special_tokens: |
|
bos_token: "<s>" |
|
eos_token: "</s>" |
|
unk_token: "<unk>" |
|
``` |
|
|
|
</details><br> |
|
|
|
# EvolCodeLlama-7b |
|
|
|
This model is a fine-tuned version of [codellama/CodeLlama-7b-hf](https://huggingface.co/codellama/CodeLlama-7b-hf) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3828 |
|
|
|
## 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: 2 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 8 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 100 |
|
- num_epochs: 3 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 0.3627 | 0.01 | 1 | 0.5027 | |
|
| 0.3412 | 0.03 | 4 | 0.5026 | |
|
| 0.3806 | 0.07 | 8 | 0.5023 | |
|
| 0.392 | 0.1 | 12 | 0.5018 | |
|
| 0.4141 | 0.14 | 16 | 0.4999 | |
|
| 0.3433 | 0.17 | 20 | 0.4954 | |
|
| 0.3702 | 0.21 | 24 | 0.4851 | |
|
| 0.2948 | 0.24 | 28 | 0.4682 | |
|
| 0.3387 | 0.28 | 32 | 0.4499 | |
|
| 0.2437 | 0.31 | 36 | 0.4331 | |
|
| 0.2526 | 0.35 | 40 | 0.4221 | |
|
| 0.2721 | 0.38 | 44 | 0.4146 | |
|
| 0.2292 | 0.42 | 48 | 0.4089 | |
|
| 0.1986 | 0.45 | 52 | 0.4028 | |
|
| 0.3258 | 0.48 | 56 | 0.3983 | |
|
| 0.3509 | 0.52 | 60 | 0.3950 | |
|
| 0.2697 | 0.55 | 64 | 0.3926 | |
|
| 0.2646 | 0.59 | 68 | 0.3907 | |
|
| 0.3979 | 0.62 | 72 | 0.3900 | |
|
| 0.2737 | 0.66 | 76 | 0.3880 | |
|
| 0.2271 | 0.69 | 80 | 0.3865 | |
|
| 0.247 | 0.73 | 84 | 0.3847 | |
|
| 0.3112 | 0.76 | 88 | 0.3824 | |
|
| 0.2724 | 0.8 | 92 | 0.3820 | |
|
| 0.207 | 0.83 | 96 | 0.3814 | |
|
| 0.3492 | 0.87 | 100 | 0.3810 | |
|
| 0.2474 | 0.9 | 104 | 0.3802 | |
|
| 0.4037 | 0.94 | 108 | 0.3785 | |
|
| 0.2295 | 0.97 | 112 | 0.3773 | |
|
| 0.2689 | 1.0 | 116 | 0.3760 | |
|
| 0.2546 | 1.02 | 120 | 0.3753 | |
|
| 0.1916 | 1.05 | 124 | 0.3768 | |
|
| 0.2458 | 1.09 | 128 | 0.3758 | |
|
| 0.2155 | 1.12 | 132 | 0.3768 | |
|
| 0.2341 | 1.16 | 136 | 0.3773 | |
|
| 0.1909 | 1.19 | 140 | 0.3793 | |
|
| 0.1911 | 1.23 | 144 | 0.3759 | |
|
| 0.2096 | 1.26 | 148 | 0.3761 | |
|
| 0.2353 | 1.29 | 152 | 0.3772 | |
|
| 0.2606 | 1.33 | 156 | 0.3773 | |
|
| 0.1485 | 1.36 | 160 | 0.3778 | |
|
| 0.1807 | 1.4 | 164 | 0.3749 | |
|
| 0.2294 | 1.43 | 168 | 0.3770 | |
|
| 0.216 | 1.47 | 172 | 0.3759 | |
|
| 0.1791 | 1.5 | 176 | 0.3727 | |
|
| 0.2605 | 1.54 | 180 | 0.3733 | |
|
| 0.2838 | 1.57 | 184 | 0.3738 | |
|
| 0.2632 | 1.61 | 188 | 0.3694 | |
|
| 0.1839 | 1.64 | 192 | 0.3686 | |
|
| 0.1939 | 1.68 | 196 | 0.3690 | |
|
| 0.2413 | 1.71 | 200 | 0.3699 | |
|
| 0.1494 | 1.74 | 204 | 0.3689 | |
|
| 0.2782 | 1.78 | 208 | 0.3695 | |
|
| 0.2314 | 1.81 | 212 | 0.3696 | |
|
| 0.2499 | 1.85 | 216 | 0.3691 | |
|
| 0.1976 | 1.88 | 220 | 0.3672 | |
|
| 0.2587 | 1.92 | 224 | 0.3660 | |
|
| 0.2598 | 1.95 | 228 | 0.3658 | |
|
| 0.2686 | 1.99 | 232 | 0.3666 | |
|
| 0.216 | 2.01 | 236 | 0.3673 | |
|
| 0.1261 | 2.04 | 240 | 0.3723 | |
|
| 0.1938 | 2.08 | 244 | 0.3811 | |
|
| 0.1906 | 2.11 | 248 | 0.3869 | |
|
| 0.1375 | 2.15 | 252 | 0.3829 | |
|
| 0.228 | 2.18 | 256 | 0.3796 | |
|
| 0.2524 | 2.22 | 260 | 0.3789 | |
|
| 0.118 | 2.25 | 264 | 0.3809 | |
|
| 0.2224 | 2.29 | 268 | 0.3834 | |
|
| 0.1477 | 2.32 | 272 | 0.3847 | |
|
| 0.2095 | 2.35 | 276 | 0.3849 | |
|
| 0.1919 | 2.39 | 280 | 0.3820 | |
|
| 0.1916 | 2.42 | 284 | 0.3804 | |
|
| 0.1625 | 2.46 | 288 | 0.3788 | |
|
| 0.2054 | 2.49 | 292 | 0.3794 | |
|
| 0.1605 | 2.53 | 296 | 0.3810 | |
|
| 0.1564 | 2.56 | 300 | 0.3819 | |
|
| 0.196 | 2.6 | 304 | 0.3822 | |
|
| 0.1975 | 2.63 | 308 | 0.3830 | |
|
| 0.1406 | 2.67 | 312 | 0.3833 | |
|
| 0.2754 | 2.7 | 316 | 0.3830 | |
|
| 0.1544 | 2.74 | 320 | 0.3829 | |
|
| 0.1733 | 2.77 | 324 | 0.3830 | |
|
| 0.1862 | 2.81 | 328 | 0.3832 | |
|
| 0.1634 | 2.84 | 332 | 0.3829 | |
|
| 0.1966 | 2.87 | 336 | 0.3830 | |
|
| 0.1306 | 2.91 | 340 | 0.3831 | |
|
| 0.1444 | 2.94 | 344 | 0.3828 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.7.2.dev0 |
|
- Transformers 4.37.0 |
|
- Pytorch 2.0.1+cu118 |
|
- Datasets 2.16.1 |
|
- Tokenizers 0.15.0 |
