language:
- en
license: mit
tags:
- code
- text-generation-inference
datasets:
- glaiveai/glaive-code-assistant-v2
- TokenBender/code_instructions_122k_alpaca_style
metrics:
- code_eval
pipeline_tag: text-generation
model-index:
- name: CodeNinja-1.0-OpenChat-7B
results:
- task:
type: text-generation
name: Text Generation
dataset:
name: IFEval (0-Shot)
type: HuggingFaceH4/ifeval
args:
num_few_shot: 0
metrics:
- type: inst_level_strict_acc and prompt_level_strict_acc
value: 54.47
name: strict accuracy
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=beowolx/CodeNinja-1.0-OpenChat-7B
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: BBH (3-Shot)
type: BBH
args:
num_few_shot: 3
metrics:
- type: acc_norm
value: 21.71
name: normalized accuracy
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=beowolx/CodeNinja-1.0-OpenChat-7B
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MATH Lvl 5 (4-Shot)
type: hendrycks/competition_math
args:
num_few_shot: 4
metrics:
- type: exact_match
value: 5.21
name: exact match
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=beowolx/CodeNinja-1.0-OpenChat-7B
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: GPQA (0-shot)
type: Idavidrein/gpqa
args:
num_few_shot: 0
metrics:
- type: acc_norm
value: 5.93
name: acc_norm
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=beowolx/CodeNinja-1.0-OpenChat-7B
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MuSR (0-shot)
type: TAUR-Lab/MuSR
args:
num_few_shot: 0
metrics:
- type: acc_norm
value: 11.54
name: acc_norm
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=beowolx/CodeNinja-1.0-OpenChat-7B
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MMLU-PRO (5-shot)
type: TIGER-Lab/MMLU-Pro
config: main
split: test
args:
num_few_shot: 5
metrics:
- type: acc
value: 22.39
name: accuracy
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=beowolx/CodeNinja-1.0-OpenChat-7B
name: Open LLM Leaderboard
CodeNinja: Your Advanced Coding Assistant
Overview
CodeNinja is an enhanced version of the renowned model openchat/openchat-3.5-1210. It having been fine-tuned through Supervised Fine Tuning on two expansive datasets, encompassing over 400,000 coding instructions. Designed to be an indispensable tool for coders, CodeNinja aims to integrate seamlessly into your daily coding routine.
Discover the quantized versions at: beowolx/CodeNinja-1.0-OpenChat-7B-GGUF.
Key Features
Expansive Training Database: CodeNinja has been refined with datasets from glaiveai/glaive-code-assistant-v2 and TokenBender/code_instructions_122k_alpaca_style, incorporating around 400,000 coding instructions across various languages including Python, C, C++, Rust, Java, JavaScript, and more.
Flexibility and Scalability: Available in a 7B model size, CodeNinja is adaptable for local runtime environments.
Advanced Code Completion: With a substantial context window size of 8192, it supports comprehensive project-level code completion.
Prompt Format
CodeNinja maintains the same prompt structure as OpenChat 3.5. Effective utilization requires adherence to this format:
GPT4 Correct User: Hello<|end_of_turn|>GPT4 Correct Assistant: Hi<|end_of_turn|>GPT4 Correct User: How are you today?<|end_of_turn|>GPT4 Correct Assistant:
🚨 Important: Ensure the use of <|end_of_turn|> as the end-of-generation token.
Adhering to this format is crucial for optimal results.
Usage Instructions
Using LM Studio
The simplest way to engage with CodeNinja is via the quantized versions on LM Studio. Ensure you select the "OpenChat" preset, which incorporates the necessary prompt format. The preset is also available in this gist.
Using the Transformers Library
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
# Initialize the model
model_path = "beowolx/CodeNinja-1.0-OpenChat-7B"
model = AutoModelForCausalLM.from_pretrained(model_path, device_map="auto")
# Load the OpenChat tokenizer
tokenizer = AutoTokenizer.from_pretrained("openchat/openchat-3.5-1210", use_fast=True)
def generate_one_completion(prompt: str):
messages = [
{"role": "user", "content": prompt},
{"role": "assistant", "content": ""} # Model response placeholder
]
# Generate token IDs using the chat template
input_ids = tokenizer.apply_chat_template(messages, add_generation_prompt=True)
# Produce completion
generate_ids = model.generate(
torch.tensor([input_ids]).to("cuda"),
max_length=256,
pad_token_id=tokenizer.pad_token_id,
eos_token_id=tokenizer.eos_token_id
)
# Process the completion
completion = tokenizer.decode(generate_ids[0], skip_special_tokens=True)
completion = completion.split("\n\n\n")[0].strip()
return completion
License
CodeNinja is licensed under the MIT License, with model usage subject to the Model License.
Contact
For queries or support, please open an issue in the repository.
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 20.21 |
| IFEval (0-Shot) | 54.47 |
| BBH (3-Shot) | 21.71 |
| MATH Lvl 5 (4-Shot) | 5.21 |
| GPQA (0-shot) | 5.93 |
| MuSR (0-shot) | 11.54 |
| MMLU-PRO (5-shot) | 22.39 |