---
language:
- en
license: other
library_name: transformers
datasets:
- Open-Orca/SlimOrca
- m-a-p/Code-Feedback
- MaziyarPanahi/WizardLM_evol_instruct_V2_196k
- camel-ai/math
- camel-ai/physics
- camel-ai/biology
- camel-ai/chemistry
- LDJnr/Capybara
- jondurbin/airoboros-3.2
- microsoft/orca-math-word-problems-200k
inference:
  parameters:
    do_sample: true
    temperature: 0.8
    top_p: 0.95
    top_k: 40
    max_new_tokens: 250
    repetition_penalty: 1.1

---

# neural-chat-mini-v2.2-1.8B

<!-- Provide a quick summary of what the model is/does. -->
We fine-tuned tau-1.8B using SFT and DPOP on a high quality mix for general-purpose assistants.


## Model Details

### Model Description

<!-- Provide a longer summary of what this model is. -->

This model has capabilities in math, coding, writing, and more. We fine-tuned it using a high quality mix for general-purpose assistants.

- **Developed by:** M4-ai
- **Language(s) (NLP):** English and maybe Chinese
- **License:** tongyi-qianwen license
- **Finetuned from model:** [tau-1.8B](https://huggingface.co/M4-ai/tau-1.8B)

## Uses

<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->

General purpose assistant, question answering, chain-of-thought, etc..

### Recommendations

<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.

## Training Details

### Training Data

- Open-Orca/SlimOrca
- m-a-p/Code-Feedback
- MaziyarPanahi/WizardLM_evol_instruct_V2_196k
- camel-ai/math
- camel-ai/physics
- camel-ai/biology
- camel-ai/chemistry
- LDJnr/Capybara
- jondurbin/airoboros-3.2
- microsoft/orca-math-word-problems-200k
- mlabonne/orpo-dpo-mix-40k

## Evaluations

coming soon

#### Training Hyperparameters

- **Training regime:** bf16 non-mixed precision
## Technical Specifications

#### Hardware

We used 8 Kaggle TPUs, and we trained at a global batch size of 128 and sequence length of 2048.