README.md

---
license: apache-2.0
tags:
  - autotrain
  - text-generation
base_model: Locutusque/TinyMistral-248M
datasets:
  - WizardLM/WizardLM_evol_instruct_V2_196k
  - KnutJaegersberg/WizardLM_evol_instruct_V2_196k_instruct_format
widget:
  - text: |-
      ### Instruction:
      Write the specs of a game about trolls and warriors in a fantasy world.

      ### Response:
  - text: |-
      ### Instruction:
      Tell me all you know about the Earth.
      
      ### Response:
  - text: |-
      ### Instruction:
      How to become more healthy?
      
      ### Response:
inference:
  parameters:
    max_new_tokens: 64
    repetition_penalty: 1.18
---

# Locutusque's TinyMistral-248M trained on the Evol Instruct dataset

- Base model: [Locutusque/TinyMistral-248M](https://huggingface.co/Locutusque/TinyMistral-248M)
- Dataset: [WizardLM/WizardLM_evol_instruct_V2_196k](https://huggingface.co/datasets/WizardLM/WizardLM_evol_instruct_V2_196k)
- Trained with [AutoTrain Advanced](https://github.com/huggingface/autotrain-advanced) using [these parameters](https://huggingface.co/Felladrin/TinyMistral-248M-Evol-Instruct/blob/321787e81e2eb0392d7ce2715154fb9c254e39b1/training_params.json) and [this CSV file](https://huggingface.co/datasets/KnutJaegersberg/WizardLM_evol_instruct_V2_196k_instruct_format/blob/93aa373501f829449f23efc91b3ac6e7a60a4d70/all_instructions.csv)
- Availability in other ML formats:
  - GGUF: [Felladrin/gguf-TinyMistral-248M-Evol-Instruct](https://huggingface.co/Felladrin/gguf-TinyMistral-248M-Evol-Instruct)

## Recommended Prompt Format

```
### Instruction:
<instruction>

### Response:
```