|
|
---
|
|
|
base_model:
|
|
|
- deepseek-ai/DeepSeek-R1-Distill-Llama-8B
|
|
|
- Skywork/Skywork-o1-Open-Llama-3.1-8B
|
|
|
- NousResearch/DeepHermes-3-Llama-3-8B-Preview
|
|
|
- O1-OPEN/OpenO1-LLama-8B-v0.1
|
|
|
- SimpleBerry/LLaMA-O1-Supervised-1129
|
|
|
- terrycraddock/Reflection-Llama-3.1-8B
|
|
|
language:
|
|
|
- en
|
|
|
tags:
|
|
|
- problem-solving
|
|
|
- chain-of-thought
|
|
|
- reasoning
|
|
|
license: llama3.1
|
|
|
---
|
|
|
|
|
|
# Llama3.1-deep-o1 |
|
|
|
|
|
This is a merge of several DeepSeek R1 distilled and O1-style long chain-of-thought (CoT) large language models (LLMs). It is designed for generating long, coherent solutions and excels at problem-solving tasks among models with 8 billion parameters. |
|
|
|
|
|
## Model Overview |
|
|
|
|
|
**Key Features:** |
|
|
- Generates detailed, manual-like explanations for complex questions. |
|
|
- Suitable for creating solution outlines, analyzing problems, and writing essays. |
|
|
|
|
|
**Limitations:** |
|
|
- Does not follow standard CoT formats like `<thought>` tags. |
|
|
- Prone to calculation errors and careless mistakes in reasoning. |
|
|
- Struggles with multiturn conversations and user alignment. |
|
|
- For example, it may ask questions mid-response but continue answering regardless, in line with its CoT origins. |
|
|
|
|
|
## Merge Details |
|
|
|
|
|
The model was created using the following YAML configuration: |
|
|
|
|
|
```yaml |
|
|
models: |
|
|
- model: deepseek-ai/DeepSeek-R1-Distill-Llama-8B |
|
|
parameters: |
|
|
weight: 1.5 |
|
|
- model: Skywork/Skywork-o1-Open-Llama-3.1-8B |
|
|
parameters: |
|
|
weight: 1.5 |
|
|
- model: NousResearch/DeepHermes-3-Llama-3-8B-Preview |
|
|
- model: O1-OPEN/OpenO1-LLama-8B-v0.1 |
|
|
- model: SimpleBerry/LLaMA-O1-Supervised-1129 |
|
|
- model: terrycraddock/Reflection-Llama-3.1-8B |
|
|
merge_method: linear |
|
|
parameters: |
|
|
weight: 1 |
|
|
dtype: bfloat16 |
|
|
``` |
|
|
|
|
|
## Usage Recommendations |
|
|
|
|
|
- To generate hints for scientific problem solving |
|
|
- As a foundation model for finetuning and merging |
|
|
|
|
|
### Examples to Try: |
|
|
- Write the equations for glycolysis and pyruvate oxidation. |
|
|
- Calculate net ATP formation from glucose metabolism (excluding electron transport chain). |
|
|
- Integrate x^2 e^x dx. |
|
|
- Prove that the complete bipartite graph K_{3,3} isn't planar. |
|
|
- Derive a formula for the critical angle between two media with refractive indices n_1 and n_2. |
|
|
- Compare steam vs. diesel engines including their capabilities and historical significance. |
|
|
|
|
|
### Notes on Performance: |
|
|
While the model provides coherent and expert-like responses, users should verify its outputs for accuracy - especially in calculations or logical reasoning tasks. |
|
|
|
|
|
## Warning |
|
|
|
|
|
- This model is experimental and may require careful validation when used for critical applications. |
|
|
- It is not optimized for conversational tasks but performs well in single-turn question answering. |
