| Model Type | | text-generation-inference, transformers, chemistry, biology, legal, art, music, finance, code, medical, climate |
|
| Use Cases |
| Areas: | | research, commercial applications |
|
| Applications: | | text generation, inference, unsloth, mistral |
|
| Primary Use Cases: | | multi-task operations, rag, function calling |
|
| Limitations: | | context window limitations |
|
| Considerations: | | Ensure ethical use and understanding of context window limitations |
|
|
| Additional Notes | | The model focuses heavily on methodology and recalling data efficiently entered into its matrix. |
|
| Supported Languages | | English (High), Swahili (Medium), Igbo (Low), Somali (Medium), Spanish (High), Catalan (Medium) |
|
| Training Details |
| Data Sources: | | gretelai/synthetic_text_to_sql, HuggingFaceTB/cosmopedia, teknium/OpenHermes-2.5, Open-Orca/SlimOrca, cognitivecomputations/dolphin-coder, databricks/databricks-dolly-15k, yonlp/CulturaX, mwitiderrick/SwahiliPlatypus |
|
| Data Volume: | |
| Methodology: | | Unsloth, Huggingface TRL, chain of thoughts, graph of thoughts, graph of thoughts, multi-task operations |
|
| Context Length: | |
| Training Time: | |
| Model Architecture: | | 32k context window, Rope-theta=1e6 |
|
|