|
|
--- |
|
|
library_name: transformers |
|
|
base_model: timarni/qwen3_dpo_100k |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
datasets: |
|
|
- timarni/MNLP_intstruction_tuning |
|
|
model-index: |
|
|
- name: outputs/dpo_100k_full_alpaca_big |
|
|
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/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl) |
|
|
<details><summary>See axolotl config</summary> |
|
|
|
|
|
axolotl version: `0.9.2` |
|
|
```yaml |
|
|
base_model: timarni/qwen3_dpo_100k |
|
|
# Automatically upload checkpoint and final model to HF |
|
|
# hub_model_id: username/custom_model_name |
|
|
|
|
|
plugins: |
|
|
- axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin |
|
|
strict: false |
|
|
|
|
|
chat_template: qwen3 |
|
|
datasets: |
|
|
- path: timarni/MNLP_intstruction_tuning # timarni/MNLP_intstruction_tuning |
|
|
type: alpaca |
|
|
split: train |
|
|
|
|
|
shuffle_merged_datasets: true |
|
|
|
|
|
val_set_size: 0.1 |
|
|
output_dir: ./outputs/dpo_100k_full_alpaca_big |
|
|
dataset_prepared_path: last_run_prepared |
|
|
|
|
|
sequence_len: 4096 #2048 |
|
|
sample_packing: true # was true -> need to check if it actually learns on the samples or not (better understand te hyperparam and event. install axolotl to debug) |
|
|
eval_sample_packing: true |
|
|
pad_to_sequence_len: true |
|
|
# train_on_inputs: true # NEW |
|
|
# group_by_length: false NEW? |
|
|
|
|
|
# To be sure that no LORA is done |
|
|
adapter: null |
|
|
lora: false |
|
|
merge_lora: false |
|
|
|
|
|
wandb_project: mnlp_project |
|
|
wandb_entity: tim-arni |
|
|
wandb_watch: |
|
|
wandb_name: dpo_100k_full_alpaca_big |
|
|
wandb_log_model: |
|
|
|
|
|
gradient_accumulation_steps: 16 |
|
|
micro_batch_size: 2 # 2 |
|
|
num_epochs: 3 |
|
|
optimizer: adamw_torch |
|
|
lr_scheduler: cosine |
|
|
learning_rate: 0.00005 # 0.00005 |
|
|
# cosine_min_lr_ratio: 0.1 |
|
|
|
|
|
warmup_ratio: 0.05 |
|
|
weight_decay: 0.01 |
|
|
|
|
|
bf16: auto |
|
|
tf32: true |
|
|
|
|
|
gradient_checkpointing: offload |
|
|
gradient_checkpointing_kwargs: |
|
|
use_reentrant: false |
|
|
resume_from_checkpoint: |
|
|
logging_steps: 1 |
|
|
gradient_clipping: 1.0 # or max_grad_norm? |
|
|
flash_attention: true |
|
|
|
|
|
evals_per_epoch: 4 |
|
|
saves_per_epoch: 2 |
|
|
save_total_limit: 25 |
|
|
special_tokens: |
|
|
|
|
|
``` |
|
|
|
|
|
</details><br> |
|
|
|
|
|
# outputs/dpo_100k_full_alpaca_big |
|
|
|
|
|
This model is a fine-tuned version of [timarni/qwen3_dpo_100k](https://huggingface.co/timarni/qwen3_dpo_100k) on the timarni/MNLP_intstruction_tuning dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.1561 |
|
|
|
|
|
## 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: 5e-05 |
|
|
- train_batch_size: 2 |
|
|
- eval_batch_size: 2 |
|
|
- seed: 42 |
|
|
- distributed_type: multi-GPU |
|
|
- num_devices: 4 |
|
|
- gradient_accumulation_steps: 16 |
|
|
- total_train_batch_size: 128 |
|
|
- total_eval_batch_size: 8 |
|
|
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
|
- lr_scheduler_type: cosine |
|
|
- lr_scheduler_warmup_steps: 13 |
|
|
- num_epochs: 3.0 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:------:|:----:|:---------------:| |
|
|
| 0.7229 | 0.0107 | 1 | 1.1382 | |
|
|
| 0.1122 | 0.2567 | 24 | 0.1830 | |
|
|
| 0.0988 | 0.5134 | 48 | 0.1736 | |
|
|
| 0.0994 | 0.7701 | 72 | 0.1662 | |
|
|
| 0.093 | 1.0214 | 96 | 0.1605 | |
|
|
| 0.07 | 1.2781 | 120 | 0.1584 | |
|
|
| 0.0685 | 1.5348 | 144 | 0.1549 | |
|
|
| 0.0695 | 1.7914 | 168 | 0.1526 | |
|
|
| 0.0657 | 2.0428 | 192 | 0.1526 | |
|
|
| 0.0504 | 2.2995 | 216 | 0.1563 | |
|
|
| 0.054 | 2.5561 | 240 | 0.1566 | |
|
|
| 0.0556 | 2.8128 | 264 | 0.1561 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.51.3 |
|
|
- Pytorch 2.5.1+cu121 |
|
|
- Datasets 3.5.1 |
|
|
- Tokenizers 0.21.1 |
|
|
|
