An open API service indexing awesome lists of open source software.

https://github.com/node0/llm-tools

My expanding collection of scripts and tools designed to aid in working with large language models, understanding their performance characteristics and context limitations.
https://github.com/node0/llm-tools

llm llm-inference statistics tokenization tokens

Last synced: 5 months ago
JSON representation

My expanding collection of scripts and tools designed to aid in working with large language models, understanding their performance characteristics and context limitations.

Awesome Lists containing this project

README

          

# Single Page Applications (SPA) Collection

This repository contains various single-page applications (SPAs) designed for specific utility tasks. Each application is a lightweight, self-contained tool implemented using HTML, CSS, and JavaScript.

---

## Github Pages Demos

#### [Tokens Per Second Simulator](https://node0.github.io/llm-tools/single-page-apps/tkps_simulator.html)

#### [Text Stats Analyzer](https://node0.github.io/llm-tools/single-page-apps/text_stats_analyzer.html)


### Local Access Instructions
___

## Applications

### 1. Text Stats Analyzer
**File:** `text_stats_analyzer.html`

The **Text Stats Analyzer** is a utility tool that allows users to analyze textual data. It provides the following features:
- Counts the number of characters in the input text.
- Calculates the number of lines in the input text.
- Estimates the number of tokens (words) based on whitespace separation.

**Usage:**
1. Open the `text_stats_analyzer.html` file in your web browser.
2. Enter or paste the text you want to analyze in the text area.
3. Click the **"Analyze Text Stats"** button to display the analysis results, including the number of characters, lines, and tokens.

**Dependencies:** None. This is a pure HTML and JavaScript-based application.

---

### 2. Tokens Per Second (tkps) Simulator
**File:** `tkps_simulator.html`

The **Tokens Per Second (tkps) Simulator** is a performance simulation tool for understanding token throughput in natural language processing (NLP) models. It allows users to:
- Input a desired tokens-per-second (tkps) rate.
- Select a predefined corpus (e.g., "Lorem Ipsum" or "A Tale of Two Cities").
- Simulate the token generation process and visualize the output.

**Usage:**
1. Open the `tkps_simulator.html` file in your web browser.
2. Specify the tkps value using the input field.
3. Choose a corpus from the dropdown menu.
4. Click the **"Prompt"** button to simulate token generation, displayed in the output text area.

**Dependencies:**
- Bootstrap CSS for styling (loaded via CDN).

---

## Getting Started

To run these applications:
1. Clone or download this repository.
2. Open the desired `.html` file in any modern web browser.

---

## Contributions

Contributions to enhance or extend the functionality of these SPAs are welcome! Please feel free to fork this repository, make changes, and submit a pull request.

---

## License

This project is licensed under the MIT License.