oh-dcft-v1-no-curation
This model is a fine-tuned version of meta-llama/Llama-3.1-8B on the mlfoundations-dev/oh-dcft-v1-no-curation-sharegpt-format dataset. It achieves the following results on the evaluation set:
- Loss: 0.5306
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-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 16
- gradient_accumulation_steps: 4
- total_train_batch_size: 512
- total_eval_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.1
- lr_scheduler_warmup_steps: 1738
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.5574 | 0.9981 | 400 | 0.5630 |
| 0.5138 | 1.9988 | 801 | 0.5373 |
| 0.4712 | 2.9944 | 1200 | 0.5306 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.3.0
- Datasets 2.21.0
- Tokenizers 0.20.2
- Downloads last month
- 1
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the model is not deployed on the HF Inference API.
Model tree for mlfoundations-dev/oh-dcft-v1-no-curation
Base model
meta-llama/Llama-3.1-8B