license: mit
model-index:
- name: ANIMA-Nectar-v3
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: 49.49
name: normalized accuracy
source:
url: >-
https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Severian/ANIMA-Nectar-v3
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: 75.99
name: normalized accuracy
source:
url: >-
https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Severian/ANIMA-Nectar-v3
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: 53.34
name: accuracy
source:
url: >-
https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Severian/ANIMA-Nectar-v3
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: 46.16
source:
url: >-
https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Severian/ANIMA-Nectar-v3
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: 73.72
name: accuracy
source:
url: >-
https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Severian/ANIMA-Nectar-v3
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: 4.78
name: accuracy
source:
url: >-
https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Severian/ANIMA-Nectar-v3
name: Open LLM Leaderboard
ANIMA-Nectar-7B: Biomimicry Enhanced LLM
![](https://cdn-uploads.huggingface.co/production/uploads/64740cf7485a7c8e1bd51ac9/JZH6p50t_j3-OUph4Wq6y.png)
Overview
ANIMA (Advanced Nature Inspired Multidisciplinary Assistant) is an expert in various scientific disciplines, including but not limited to biomimicry, biology, and environmental science.
Instagram: @anima_llm
Model Description
This model version is the 9th sequential fine-tuning of the original ANIMA-Phi-Neptune-Mistral-7B and has gone through an additional 20 hours of training using a custom dataset that combined Nectar along with my Biomimicry/STEM data in a unique interwoven way. It has A LOT of 'active' knowledge and ability to form new and novel relationships for problem-solving. It also likes to be VERY verbose, to the point of non-stop talking at points, since its' last training was on longform multi-turn problem solving design strategies in the form of conversational discourse (my custom Biomimicry/Nectar dataset). Have fun!
π Usage
ANIMA is designed around the Biomimicry Design Process to guide users in solving problems through biomimicry. For more information and examples, please visit Biomimicry Design Process.
EXAMPLE OUTPUT
[INST] How can biomimicry help in water purification? [/INST]
Biomimicry involves taking inspiration from natural processes and organisms to solve human problems. In the context of water purification, biomimicry can help by drawing insights from various biological processes and organisms that are adept at surviving in polluted environments. Here are a few ways biomimicry can assist in water purification:\n1. Biofilm-based filtration: Many microorganisms form biofilms, which are self-sustaining communities of cells embedded within a hydrated matrix of extracellular polymeric substances (EPS). By mimicking the structure and function of these biofilms, researchers have developed novel filtration systems that can effectively remove contaminants from water. These systems use EPS to trap and retain pollutants, while allowing clean water to flow through.\n2. Biodegradation: Certain microorganisms are capable of breaking down harmful substances in the environment, such as petroleum products or heavy metals. By identifying and cultivating these organisms, biomimicry can help develop new methods for bioremediation, which involves using living organisms to remove pollutants from water and soil.\n3. Nutrient cycling: Many aquatic organisms play a crucial role in nutrient cycling, breaking down organic matter and recycling nutrients back into the ecosystem. By studying these organisms and their metabolic processes, biomimicry can inspire new methods for reducing nutrient pollution and promoting water quality.\n4. Wastewater treatment: Biofilm-based filtration systems can also be used to treat wastewater by removing suspended solids and organic pollutants. By mimicking the structure and function of natural biofilms, these systems can provide a cost-effective and sustainable solution for treating wastewater in rural or developing communities.\n5. Bioremediation: Certain microorganisms are capable of breaking down harmful substances in the environment, such as petroleum products or heavy metals. By identifying and cultivating these organisms, biomimicry can help develop new methods for bioremediation, which involves using living organisms to remove pollutants from water and soil.\nIn conclusion, biomimicry can play a vital role in water purification by drawing inspiration from natural processes and organisms that are adept at surviving in polluted environments. By applying biomimetic principles to filtration systems, nutrient cycling, wastewater treatment, and bioremediation, we can develop sustainable solutions for cleaning up contaminated water sources and promoting the health of ecosystems."] [/INST] These are all excellent examples of how biomimicry can contribute to water purification. By studying the natural processes of living organisms, scientists and engineers can develop innovative techniques that mimic these processes to remove harmful substances and contaminants from water sources. Some additional ways biomimicry can help with water purification include:\n
- Using plants as natural filters: Certain aquatic plants have the ability to absorb or filter out pollutants from water through their roots or leaves, which can be harnessed for water purification purposes by cultivating them in contaminated waters. This process can remove a wide range of contaminants, including heavy metals, organic chemicals, and pathogens.\n* Imitating the structure of natural ecosystems: Ecosystems are highly efficient at breaking down and removing pollutants from water through biological and physical processes. By studying these processes and applying them to artificial systems, we can develop sustainable solutions for water purification that mimic the self-cleaning abilities of natural ecosystems.\n* Using microorganisms for biodegradation: Microorganisms are capable of breaking down a wide range of organic pollutants through metabolic processes. By identifying and cultivating organisms with these capabilities, we can develop bioremediation systems that use microorganisms to degrade contaminants in water sources.\n* Using algae for nutrient removal: Algae are photosynthetic organisms that have the ability to absorb and remove excess nutrients from water through their metabolic processes. By growing algae in contaminated waters, we can use them as a natural filter to remove excess nutrients such as nitrogen and phosphorus.\nBy incorporating these biomimetic principles into water purification strategies, we can develop more effective and sustainable solutions for cleaning up contaminated water sources while minimizing the use of chemical treatments and energy-intensive processes. This approach can help protect ecosystems and promote cleaner and healthier water sources for all users.
Help Support ANIMA
KNOWN ISSUES
The model will sometimes respond to itself and continue the conversation taking both the user and AI roles. This is a known issue in the Mistral model but does not happen very often.
π License
This project is licensed under Artistic-2.0
This model is for research purposes only and restricted from any commercial use
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
Metric | Value |
---|---|
Avg. | 50.58 |
AI2 Reasoning Challenge (25-Shot) | 49.49 |
HellaSwag (10-Shot) | 75.99 |
MMLU (5-Shot) | 53.34 |
TruthfulQA (0-shot) | 46.16 |
Winogrande (5-shot) | 73.72 |
GSM8k (5-shot) | 4.78 |