ruadapt_qwen2.5_3B_ext_u48_full_lr3e4_peft_bs256_v2
This model is a fine-tuned version of RefalMachine/ruadapt_qwen2.5_3B_ext_u48_mean_init on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.4070
- Accuracy: 0.5068
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: 0.0003
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 64
- gradient_accumulation_steps: 2
- total_train_batch_size: 256
- total_eval_batch_size: 128
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-05
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- num_epochs: 1.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 0.0001 | 1 | 5.8817 | 0.3107 |
| 2.5884 | 0.1765 | 2000 | 2.4509 | 0.5014 |
| 2.5525 | 0.3531 | 4000 | 2.4229 | 0.5046 |
| 2.5286 | 0.5296 | 6000 | 2.4123 | 0.5061 |
| 2.53 | 0.7062 | 8000 | 2.4078 | 0.5068 |
| 2.528 | 0.8827 | 10000 | 2.4070 | 0.5067 |
Framework versions
- PEFT 0.9.0
- Transformers 4.45.2
- Pytorch 2.3.0a0+6ddf5cf85e.nv24.04
- Datasets 2.18.0
- Tokenizers 0.20.1
- Downloads last month
- 4