Dolly V2 12B is an open-source language model by databricks. Features: 12b LLM, VRAM: 23.8GB, Context: 2K, License: mit, LLM Explorer Score: 0.27, ELO: 982, Arc: 42.4, HellaSwag: 72.5, MMLU: 25.9, GSM8K: 1.2.
Dolly V2 12B Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Dolly V2 12B Parameters and Internals
| Model Type | |
| Use Cases |
| Primary Use Cases: | |
| Limitations: | | syntactically complex prompts, programming problems, mathematical operations, factual errors, dates and times, open-ended question answering, hallucination, enumerating lists of specific length, stylistic mimicry, humor, well-formatted letter writing |
|
|
| Training Details |
| Data Sources: | | EleutherAI Pythia, databricks-dolly-15k |
|
| Methodology: | |
| Model Architecture: | | Derived from EleutherAI's Pythia-12b |
|
|
| Input Output |
| Accepted Modalities: | |
| Performance Tips: | | Include `torch_dtype=torch.bfloat16` to reduce memory usage if supported. Use `trust_remote_code=True` when loading the pipeline. |
|
|
Best Alternatives to Dolly V2 12B