metadata

model-index:
  - name: sparrow-1.1b-chat-alpha
    results: []
license: apache-2.0
inference: false
datasets:
  - stingning/ultrachat
  - OpenAssistant/oasst1
language:
  - en

Model Card for Sparrow 1.1B Chat Alpha GGUF

The Sparrow series comprises language models designed to serve as helpful assistants and as a base model for domain-specific fine tuning. Sparrow-1.1B-Chat-α is the initial model in this series and represents a fine-tuned iteration of PY007/TinyLlama-1.1B-intermediate-step-480k-1T. It was trained on a combination of publicly accessible and synthetic datasets.

Model Description:

Model Type: Sparrow-7B-α is a 1.1B parameter model, that has been fine-tuned using a mixture of publicly available and synthetic datasets.
Supported Languages (NLP): The primary language is English.
License/Warranty: The model is available under the Apache 2.0 license and comes with no warranty or gurantees of any kind.
Fine-tuned from: PY007/TinyLlama-1.1B-intermediate-step-480k-1T

Repo contains models converted to GGUF either in int8 or float16 format:

sparrow-1.1b-chat-alpha-q8_0.gguf
sparrow-1.1b-chat-alpha-f16.gguf

Prompt Template:

<s>[INST] <<SYS>>
{{ system_prompt }}
<</SYS>>

{{ user_message }} [/INST]