zlm-ceb_b64_le5_s8000
This model is a fine-tuned version of mikhail-panzo/zlm_b64_le4_s12000 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4051
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: 0.0001
- train_batch_size: 32
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2000
- training_steps: 8000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.4626 | 19.6078 | 500 | 0.4263 |
| 0.4288 | 39.2157 | 1000 | 0.4077 |
| 0.4109 | 58.8235 | 1500 | 0.4013 |
| 0.3978 | 78.4314 | 2000 | 0.4035 |
| 0.3898 | 98.0392 | 2500 | 0.4013 |
| 0.373 | 117.6471 | 3000 | 0.4010 |
| 0.3644 | 137.2549 | 3500 | 0.4005 |
| 0.3569 | 156.8627 | 4000 | 0.4029 |
| 0.3515 | 176.4706 | 4500 | 0.4039 |
| 0.3443 | 196.0784 | 5000 | 0.4005 |
| 0.3469 | 215.6863 | 5500 | 0.4018 |
| 0.3427 | 235.2941 | 6000 | 0.4001 |
| 0.3401 | 254.9020 | 6500 | 0.4042 |
| 0.3419 | 274.5098 | 7000 | 0.4054 |
| 0.3318 | 294.1176 | 7500 | 0.4057 |
| 0.3312 | 313.7255 | 8000 | 0.4051 |
Framework versions
- Transformers 4.41.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1
- Downloads last month
- 0