Skip to content

models databricks dolly v2 12b

github-actions[bot] edited this page Jun 2, 2023 · 27 revisions

databricks-dolly-v2-12b

Overview

Description: The dolly-v2-12b model is a large, instruction-following language model created by Databricks, Inc. and is licensed for commercial use. It is based on EleutherAI’s Pythia-12b and was trained on a 15,000-record instruction corpus generated by Databricks employees, which was released under a CC-BY-SA license. It is not a state-of-the-art model, but it does exhibit high-quality instruction-following behavior that is not characteristic of the foundation model on which it is based. > The above summary was generated using ChatGPT. Review the original model card to understand the data used to train the model, evaluation metrics, license, intended uses, limitations and bias before using the model. > NOTE: It is recommended to use Standard_NC24rs_v3 SKU and set the below mentioned parameters in the advanced wizard when launching a finetune job. > - Precision to 16 > - Apply Deepspeed to true > - Train batch size to 1 > - Validation batch size to 1 ### Finetuning samples Task|Use case|Dataset|Python sample (Notebook)|CLI with YAML |--|--|--|--|--| Text Classification|Emotion Detection|Emotion|emotion-detection.ipynb|emotion-detection.sh Token Classification|Named Entity Recognition|Conll2003|named-entity-recognition.ipynb|named-entity-recognition.sh Question Answering|Extractive Q&A|SQUAD (Wikipedia)|extractive-qa.ipynb|extractive-qa.sh ### Model Evaluation Task| Use case| Dataset| Python sample (Notebook)| CLI with YAML |--|--|--|--|--| Text generation | Text generation | cnn_dailymail | evaluate-model-text-generation.ipynb | evaluate-model-text-generation.yml ### Sample inputs and outputs (for real-time inference) json { "input_data": { "input_string": ["Hello! How are you?" , "Explain to me the difference between nuclear fission and fusion."] } } #### Sample output json [ "As a Large Language Model (LLM), I don't have feelings like humans do. But I can say that I will output this response tomorrow when I'm ready.", "Nuclear fission and fusion are different methods of releasing energy from nuclear reactions. Nuclear fission involves splitting an atomic nucleus and releasing two or more smaller atomic nuclei and nuclear fusion is a type of nuclear reaction in which two atomic nuclei merge to form a bigger atomic nucleus and release energy." ]

Version: 3

Tags

Featured Preview license : apache-2.0 task : text-generation

View in Studio: https://ml.azure.com/registries/azureml/models/databricks-dolly-v2-12b/version/3

License: apache-2.0

Properties

SHA: d0aa7ea43dc3548a8b499cc88605e150c4a2704d

datasets: databricks/databricks-dolly-15k

evaluation-min-sku-spec: 6|0|56|112

evaluation-recommended-sku: Standard_DS5_v2

inference-min-sku-spec: 6|0|56|112

inference-recommended-sku: Standard_DS5_v2

finetune-min-sku-spec: 4|1|28|176

finetune-recommended-sku: Standard_ND40RS_v2

finetuning-tasks: text-classification, token-classification, question-answering

languages: en

Clone this wiki locally