mistral_darulm_20_05_24_part1-2_32000_bpe_part1-2_lr1e4_bs256
This model is a fine-tuned version of RefalMachine/mistral_darulm_20_05_24_part1-2_32000_bpe_mean_init_03_07_24 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.0469
- Accuracy: 0.5649
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: 4
- eval_batch_size: 4
- seed: 42
- distributed_type: multi-GPU
- num_devices: 64
- total_train_batch_size: 256
- total_eval_batch_size: 256
- 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 |
|---|---|---|---|---|
| 2.3626 | 0.08 | 2000 | 2.1568 | 0.5493 |
| 2.3318 | 0.17 | 4000 | 2.1260 | 0.5534 |
| 2.3106 | 0.25 | 6000 | 2.1091 | 0.5558 |
| 2.2928 | 0.33 | 8000 | 2.0945 | 0.5573 |
| 2.2619 | 0.42 | 10000 | 2.0814 | 0.5598 |
| 2.2682 | 0.5 | 12000 | 2.0695 | 0.5615 |
| 2.2519 | 0.58 | 14000 | 2.0593 | 0.5629 |
| 2.2474 | 0.67 | 16000 | 2.0529 | 0.5641 |
| 2.2326 | 0.75 | 18000 | 2.0487 | 0.5645 |
| 2.2487 | 0.83 | 20000 | 2.0472 | 0.5648 |
| 2.2439 | 0.92 | 22000 | 2.0469 | 0.5649 |
| 2.2466 | 1.0 | 24000 | 2.0469 | 0.5649 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.3.0a0+6ddf5cf85e.nv24.04
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 5
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.