The Trial - Initial commit
This commit is contained in:
45
README.md
Normal file
45
README.md
Normal file
@@ -0,0 +1,45 @@
|
||||
# The Trial Literary Analysis SLM
|
||||
|
||||
## Project Overview
|
||||
Building a specialized Small Language Model (SLM) for comprehensive analysis of "The Trial" by Franz Kafka.
|
||||
|
||||
## Capabilities
|
||||
- Factual Q&A about plot, characters, and timeline
|
||||
- Literary analysis (themes, symbolism, narrative techniques)
|
||||
- Creative content generation in Kafka's style
|
||||
- Contextual conversation with cross-references
|
||||
|
||||
## Hardware Requirements
|
||||
- GPU: 8-16GB VRAM (RTX 3080/4080 recommended)
|
||||
- Storage: 50GB for models and data
|
||||
- RAM: 16GB+ recommended
|
||||
|
||||
## Project Structure
|
||||
```
|
||||
the-trial-slm/
|
||||
├── data/ # Training datasets
|
||||
├── models/ # Model files and adapters
|
||||
├── scripts/ # Training and utility scripts
|
||||
├── notebooks/ # Jupyter notebooks for development
|
||||
├── tests/ # Evaluation and testing scripts
|
||||
└── deployment/ # Ollama integration files
|
||||
```
|
||||
|
||||
## Training Phases
|
||||
1. **Data Preparation** (Weeks 1-2)
|
||||
2. **Infrastructure Setup** (Week 3)
|
||||
3. **Model Training** (Weeks 4-5)
|
||||
4. **Ollama Integration** (Week 6)
|
||||
5. **Testing & Refinement** (Week 7)
|
||||
|
||||
## Base Model
|
||||
- **Primary**: Llama 3.2 3B Instruct
|
||||
- **Method**: QLoRA fine-tuning
|
||||
- **Target**: Consumer GPU deployment
|
||||
|
||||
## Key Differences from Monte Cristo Project
|
||||
- **Source Text**: "The Trial" by Franz Kafka (Project Gutenberg #7849)
|
||||
- **Literary Period**: Early 20th century existential literature
|
||||
- **Key Themes**: Bureaucratic absurdity, alienation, guilt vs innocence
|
||||
- **Style**: Absurdist, nightmarish, psychological realism
|
||||
- **Protagonist**: Josef K. (bank clerk arrested without cause)
|
||||
Reference in New Issue
Block a user