66 lines
2.6 KiB
Markdown
66 lines
2.6 KiB
Markdown
|
---
|
||
|
|
base_model: Dimensity/Dimensity-3B
|
||
|
|
inference: false
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
license: mit
|
||
|
|
model_creator: Dimensity
|
||
|
|
model_name: Dimensity-3B
|
||
|
|
pipeline_tag: text-generation
|
||
|
|
quantized_by: afrideva
|
||
|
|
tags:
|
||
|
|
- sft
|
||
|
|
- gguf
|
||
|
|
- ggml
|
||
|
|
- quantized
|
||
|
|
- q2_k
|
||
|
|
- q3_k_m
|
||
|
|
- q4_k_m
|
||
|
|
- q5_k_m
|
||
|
|
- q6_k
|
||
|
|
- q8_0
|
||
|
|
---
|
||
|
|
# Dimensity/Dimensity-3B-GGUF
|
||
|
|
|
||
|
|
Quantized GGUF model files for [Dimensity-3B](https://huggingface.co/Dimensity/Dimensity-3B) from [Dimensity](https://huggingface.co/Dimensity)
|
||
|
|
|
||
|
|
|
||
|
|
| Name | Quant method | Size |
|
||
|
|
| ---- | ---- | ---- |
|
||
|
|
| [dimensity-3b.fp16.gguf](https://huggingface.co/afrideva/Dimensity-3B-GGUF/resolve/main/dimensity-3b.fp16.gguf) | fp16 | 5.59 GB |
|
||
|
|
| [dimensity-3b.q2_k.gguf](https://huggingface.co/afrideva/Dimensity-3B-GGUF/resolve/main/dimensity-3b.q2_k.gguf) | q2_k | 1.20 GB |
|
||
|
|
| [dimensity-3b.q3_k_m.gguf](https://huggingface.co/afrideva/Dimensity-3B-GGUF/resolve/main/dimensity-3b.q3_k_m.gguf) | q3_k_m | 1.39 GB |
|
||
|
|
| [dimensity-3b.q4_k_m.gguf](https://huggingface.co/afrideva/Dimensity-3B-GGUF/resolve/main/dimensity-3b.q4_k_m.gguf) | q4_k_m | 1.71 GB |
|
||
|
|
| [dimensity-3b.q5_k_m.gguf](https://huggingface.co/afrideva/Dimensity-3B-GGUF/resolve/main/dimensity-3b.q5_k_m.gguf) | q5_k_m | 1.99 GB |
|
||
|
|
| [dimensity-3b.q6_k.gguf](https://huggingface.co/afrideva/Dimensity-3B-GGUF/resolve/main/dimensity-3b.q6_k.gguf) | q6_k | 2.30 GB |
|
||
|
|
| [dimensity-3b.q8_0.gguf](https://huggingface.co/afrideva/Dimensity-3B-GGUF/resolve/main/dimensity-3b.q8_0.gguf) | q8_0 | 2.97 GB |
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
## Original Model Card:
|
||
|
|
```Dimensity-3B```
|
||
|
|
|
||
|
|
# Model Details
|
||
|
|
Dimensity-3B is a finetuned version of the StableLM framework trained on a variety of conversational data. It contains 3 billion parameters.
|
||
|
|
|
||
|
|
# Intended Uses
|
||
|
|
This model is intended for conversational AI applications. It can engage in open-ended dialogue by generating responses to user prompts.
|
||
|
|
|
||
|
|
## Factors
|
||
|
|
# Training Data
|
||
|
|
The model was trained on a large dataset of over 100 million conversational exchanges extracted from Reddit comments, customer support logs, and other online dialogues.
|
||
|
|
|
||
|
|
# Prompt Template
|
||
|
|
The model was finetuned using the following prompt template:
|
||
|
|
```
|
||
|
|
### Human: {prompt}
|
||
|
|
|
||
|
|
### Assistant:
|
||
|
|
```
|
||
|
|
This prompts the model to take on an assistant role.
|
||
|
|
|
||
|
|
# Ethical Considerations
|
||
|
|
As the model was trained on public conversational data, it may generate responses that contain harmful stereotypes or toxic content. The model should be used with caution in sensitive contexts.
|
||
|
|
|
||
|
|
# Caveats and Recommendations
|
||
|
|
This model is designed for open-ended conversation. It may sometimes generate plausible-sounding but incorrect information. Outputs should be validated against external sources.
|