metadata
language:
- en
license: apache-2.0
tags:
- nlp
- language-model
- deepseek
- openslm
OpenSLM Deepseek 30M
This is the Deepseek model trained with approximately 30M parameters, as part of the OpenSLM project.
Model Details
- Architecture: Deepseek
- Parameters: ~30M
- License: Apache 2.0
- Language: English
Performance
Based on the evaluation metrics, the model achieved the following performance:
- Best Validation Loss:
3.1681
Usage
For detailed usage instructions, please refer to the OpenSLM repository.