106 lines
3.9 KiB
Markdown
106 lines
3.9 KiB
Markdown
---
|
|
library_name: transformers
|
|
tags:
|
|
- text-generation-inference
|
|
- code
|
|
- math
|
|
- sft
|
|
license: apache-2.0
|
|
language:
|
|
- en
|
|
base_model:
|
|
- Qwen/Qwen2.5-1.5B-Instruct
|
|
pipeline_tag: text-generation
|
|
---
|
|
|
|
|
|
|
|
# **Lambda-Cancri-1.5B-Exp**
|
|
|
|
> **Lambda-Cancri-1.5B-Exp** is a **general-purpose LLM** fine-tuned from **Qwen2.5-1.5B** using **optimized supervised fine-tuning (SFT)**. This model is designed to enhance coding proficiency, reasoning ability, and step-by-step explanation across multiple software development and general knowledge tasks.
|
|
|
|
## **Key Features**
|
|
|
|
1. **Code Reasoning & Explanation**
|
|
Trained to **analyze, generate, and explain** code with a focus on logic, structure, and clarity. Supports functional, object-oriented, and procedural paradigms.
|
|
|
|
2. **Optimized Supervised Fine-Tuning (SFT)**
|
|
Fine-tuned using **high-quality supervised datasets**, ensuring strong performance in tasks like code generation, bug fixing, function completion, and abstract reasoning.
|
|
|
|
3. **Multi-Language & General Task Support**
|
|
Works fluently with **Python**, **JavaScript**, **C++**, **Shell**, and general knowledge tasks — ideal for programming, scripting, algorithmic problem-solving, and general-purpose reasoning.
|
|
|
|
4. **Compact and Efficient**
|
|
At just **1.5B parameters**, it's lightweight enough for **edge deployments**, **developer tools**, and **general-purpose assistants**, while maintaining strong reasoning and coding capabilities.
|
|
|
|
5. **Debugging and Auto-Fix Capabilities**
|
|
Built to **identify bugs**, **recommend corrections**, and provide **context-aware explanations** of issues across a wide range of codebases.
|
|
|
|
## **Quickstart with Transformers**
|
|
|
|
```python
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer
|
|
|
|
model_name = "prithivMLmods/Lambda-Cancri-1.5B-Exp"
|
|
|
|
model = AutoModelForCausalLM.from_pretrained(
|
|
model_name,
|
|
torch_dtype="auto",
|
|
device_map="auto"
|
|
)
|
|
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
|
|
|
prompt = "Write a Python function that checks if a number is prime, and explain how it works."
|
|
|
|
messages = [
|
|
{"role": "system", "content": "You are a helpful coding and reasoning assistant. Your job is to write correct code and explain the logic step-by-step."},
|
|
{"role": "user", "content": prompt}
|
|
]
|
|
text = tokenizer.apply_chat_template(
|
|
messages,
|
|
tokenize=False,
|
|
add_generation_prompt=True
|
|
)
|
|
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
|
|
|
|
generated_ids = model.generate(
|
|
**model_inputs,
|
|
max_new_tokens=512
|
|
)
|
|
generated_ids = [
|
|
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
|
|
]
|
|
|
|
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
|
|
```
|
|
|
|
## **Intended Use**
|
|
|
|
- **Code Assistance & IDE Integration**:
|
|
Smart autocomplete, bug detection, function suggestion, and explanations for developers.
|
|
|
|
- **Learning & Education**:
|
|
Perfect for **students**, **educators**, and **self-learners** in programming and technical fields.
|
|
|
|
- **Automated Code Review & QA**:
|
|
Assists in logic analysis, structure evaluation, and bug spotting in code for quality assurance.
|
|
|
|
- **General Purpose Assistant**:
|
|
Provides help beyond coding—answering general queries, solving reasoning tasks, and assisting in workflows.
|
|
|
|
- **Edge & DevTool Deployments**:
|
|
Lightweight for browser extensions, desktop applications, and CLI-based assistants.
|
|
|
|
## **Limitations**
|
|
|
|
1. **Scaling Challenges**
|
|
May not handle extremely large or highly complex projects as well as larger models.
|
|
|
|
2. **Creativity Variability**
|
|
May show inconsistent performance on highly creative or unconventional coding tasks.
|
|
|
|
3. **Security Considerations**
|
|
Outputs should be audited to ensure the generation of secure, safe code.
|
|
|
|
4. **Instruction Sensitivity**
|
|
Responds better with clear, structured prompts and task instructions. |