metadata
license: mit
0506_9_8
This model is a fine-tuned version of ../../models/Qwen1.5-7B-sft-0502 on the alpaca_formatted_review_new_data_0505_greater_8 dataset. It achieves the following results on the evaluation set:
- Loss: 0.5497
Model description
Qwen1.5 is the beta version of Qwen2, a transformer-based decoder-only language model pretrained on a large amount of data. In comparison with the previous released Qwen, the improvements include:
- 8 model sizes, including 0.5B, 1.8B, 4B, 7B, 14B, 32B and 72B dense models, and an MoE model of 14B with 2.7B activated;
- Significant performance improvement in Chat models;
- Multilingual support of both base and chat models;
- Stable support of 32K context length for models of all sizes
- No need of
trust_remote_code
.
For more details, please refer to the blog post and GitHub repo.
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: 7e-05
- train_batch_size: 2
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- total_eval_batch_size: 2
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 13
- num_epochs: 5.0
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.6358 | 0.7619 | 20 | 0.5865 |
0.6379 | 1.5238 | 40 | 0.5621 |
0.6067 | 2.2857 | 60 | 0.5561 |
0.5339 | 3.0476 | 80 | 0.5515 |
0.6749 | 3.8095 | 100 | 0.5500 |
0.6351 | 4.5714 | 120 | 0.5497 |
Framework versions
- PEFT 0.10.0
- Transformers 4.40.0
- Pytorch 2.1.0+cu121
- Datasets 2.14.5
- Tokenizers 0.19.1