199 lines
6.2 KiB
Markdown
199 lines
6.2 KiB
Markdown
---
|
|
language:
|
|
- en
|
|
license: apache-2.0
|
|
tags:
|
|
- story
|
|
- young children
|
|
- educational
|
|
- knowledge
|
|
base_model: mistralai/Mistral-7B-v0.1
|
|
datasets:
|
|
- ajibawa-2023/Children-Stories-Collection
|
|
model-index:
|
|
- name: Young-Children-Storyteller-Mistral-7B
|
|
results:
|
|
- task:
|
|
type: text-generation
|
|
name: Text Generation
|
|
dataset:
|
|
name: AI2 Reasoning Challenge (25-Shot)
|
|
type: ai2_arc
|
|
config: ARC-Challenge
|
|
split: test
|
|
args:
|
|
num_few_shot: 25
|
|
metrics:
|
|
- type: acc_norm
|
|
value: 68.69
|
|
name: normalized accuracy
|
|
source:
|
|
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ajibawa-2023/Young-Children-Storyteller-Mistral-7B
|
|
name: Open LLM Leaderboard
|
|
- task:
|
|
type: text-generation
|
|
name: Text Generation
|
|
dataset:
|
|
name: HellaSwag (10-Shot)
|
|
type: hellaswag
|
|
split: validation
|
|
args:
|
|
num_few_shot: 10
|
|
metrics:
|
|
- type: acc_norm
|
|
value: 84.67
|
|
name: normalized accuracy
|
|
source:
|
|
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ajibawa-2023/Young-Children-Storyteller-Mistral-7B
|
|
name: Open LLM Leaderboard
|
|
- task:
|
|
type: text-generation
|
|
name: Text Generation
|
|
dataset:
|
|
name: MMLU (5-Shot)
|
|
type: cais/mmlu
|
|
config: all
|
|
split: test
|
|
args:
|
|
num_few_shot: 5
|
|
metrics:
|
|
- type: acc
|
|
value: 64.11
|
|
name: accuracy
|
|
source:
|
|
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ajibawa-2023/Young-Children-Storyteller-Mistral-7B
|
|
name: Open LLM Leaderboard
|
|
- task:
|
|
type: text-generation
|
|
name: Text Generation
|
|
dataset:
|
|
name: TruthfulQA (0-shot)
|
|
type: truthful_qa
|
|
config: multiple_choice
|
|
split: validation
|
|
args:
|
|
num_few_shot: 0
|
|
metrics:
|
|
- type: mc2
|
|
value: 62.62
|
|
source:
|
|
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ajibawa-2023/Young-Children-Storyteller-Mistral-7B
|
|
name: Open LLM Leaderboard
|
|
- task:
|
|
type: text-generation
|
|
name: Text Generation
|
|
dataset:
|
|
name: Winogrande (5-shot)
|
|
type: winogrande
|
|
config: winogrande_xl
|
|
split: validation
|
|
args:
|
|
num_few_shot: 5
|
|
metrics:
|
|
- type: acc
|
|
value: 81.22
|
|
name: accuracy
|
|
source:
|
|
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ajibawa-2023/Young-Children-Storyteller-Mistral-7B
|
|
name: Open LLM Leaderboard
|
|
- task:
|
|
type: text-generation
|
|
name: Text Generation
|
|
dataset:
|
|
name: GSM8k (5-shot)
|
|
type: gsm8k
|
|
config: main
|
|
split: test
|
|
args:
|
|
num_few_shot: 5
|
|
metrics:
|
|
- type: acc
|
|
value: 65.2
|
|
name: accuracy
|
|
source:
|
|
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ajibawa-2023/Young-Children-Storyteller-Mistral-7B
|
|
name: Open LLM Leaderboard
|
|
---
|
|
|
|
|
|
**Young-Children-Storyteller-Mistral-7B**
|
|
|
|
This model is based on my dataset [Children-Stories-Collection](https://huggingface.co/datasets/ajibawa-2023/Children-Stories-Collection) which has over 0.9 million stories meant for Young Children (age 6 to 12).
|
|
|
|
Drawing upon synthetic datasets meticulously designed with the developmental needs of young children in mind, Young-Children-Storyteller is more than just a tool—it's a companion on the journey of discovery and learning.
|
|
With its boundless storytelling capabilities, this model serves as a gateway to a universe brimming with wonder, adventure, and endless possibilities.
|
|
|
|
Whether it's embarking on a whimsical adventure with colorful characters, unraveling mysteries in far-off lands, or simply sharing moments of joy and laughter, Young-Children-Storyteller fosters a love for language and storytelling from the earliest of ages.
|
|
Through interactive engagement and age-appropriate content, it nurtures creativity, empathy, and critical thinking skills, laying a foundation for lifelong learning and exploration.
|
|
|
|
Rooted in a vast repository of over 0.9 million specially curated stories tailored for young minds, Young-Children-Storyteller is poised to revolutionize the way children engage with language and storytelling.
|
|
|
|
Kindly note this is qLoRA version, another exception.
|
|
|
|
|
|
**GGUF & Exllama**
|
|
|
|
Standard Q_K & GGUF: [Link](https://huggingface.co/MarsupialAI/Young-Children-Storyteller-Mistral-7B_iMatrix_GGUF/tree/main)
|
|
|
|
Exllama: TBA
|
|
|
|
Special Thanks to [MarsupialAI](https://huggingface.co/MarsupialAI) for quantizing the model.
|
|
|
|
**Training**
|
|
|
|
Entire dataset was trained on 4 x A100 80GB. For 3 epoch, training took more than 30 Hours. Axolotl codebase was used for training purpose. Entire data is trained on Mistral-7B-v0.1.
|
|
|
|
**Example Prompt:**
|
|
|
|
This model uses **ChatML** prompt format.
|
|
|
|
```
|
|
<|im_start|>system
|
|
You are a Helpful Assistant who can write educational stories for Young Children.<|im_end|>
|
|
<|im_start|>user
|
|
{prompt}<|im_end|>
|
|
<|im_start|>assistant
|
|
|
|
```
|
|
You can modify above Prompt as per your requirement.
|
|
|
|
|
|
I want to say special Thanks to the Open Source community for helping & guiding me to better understand the AI/Model development.
|
|
|
|
Thank you for your love & support.
|
|
|
|
**Example Output**
|
|
|
|
Example 1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Example 2
|
|
|
|
|
|
|
|
|
|
|
|
Example 3
|
|
|
|
|
|
|
|
|
|
|
|
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
|
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_ajibawa-2023__Young-Children-Storyteller-Mistral-7B)
|
|
|
|
| Metric |Value|
|
|
|---------------------------------|----:|
|
|
|Avg. |71.08|
|
|
|AI2 Reasoning Challenge (25-Shot)|68.69|
|
|
|HellaSwag (10-Shot) |84.67|
|
|
|MMLU (5-Shot) |64.11|
|
|
|TruthfulQA (0-shot) |62.62|
|
|
|Winogrande (5-shot) |81.22|
|
|
|GSM8k (5-shot) |65.20|
|
|
|