RichardErkhov/raidhon_-_coven_tiny_1.1b_32k_orpo_alpha-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

coventiny1.1b32korpo_alpha - GGUF

• —Model creator: https://huggingface.co/raidhon/
• —Original model: https://huggingface.co/raidhon/coventiny1.1b32korpo_alpha/

Original model description: --- language:

• —en license: apache-2.0 tags:
• —text-generation
• —large-language-model
• —orpo dataset:
• —jondurbin/truthy-dpo-v0.1
• —AlekseyKorshuk/evol-codealpaca-v1-dpo
• —argilla/distilabel-intel-orca-dpo-pairs
• —argilla/ultrafeedback-binarized-avg-rating-for-dpo-filtered
• —snorkelai/Snorkel-Mistral-PairRM-DPO-Dataset
• —mlabonne/orpo-dpo-mix-40k

base_model:

• —TinyLlama/TinyLlama-1.1B-Chat-v1.0 model-index:
• —name: Coven Tiny 1.1B description: "Coven Tiny 1.1B is a derivative of TinyLlama 1.1B Chat, fine-tuned to perform specialized tasks involving deeper understanding and reasoning over context. This model exhibits strong capabilities in both general language understanding and task-specific challenges." results:
• —task: type: text-generation name: Winogrande Challenge dataset: name: Winogrande type: winogrande config: winograndexl split: test args: numfew_shot: 5 metrics:
• —type: accuracy value: 61.17 name: accuracy
• —task: type: text-generation name: TruthfulQA Generation dataset: name: TruthfulQA type: truthfulqa config: multiplechoice split: validation args: numfewshot: 0 metrics:
• —type: accuracy value: 34.31 name: accuracy
• —task: type: text-generation name: PIQA Problem Solving dataset: name: PIQA type: piqa split: validation args: numfewshot: 5 metrics:
• —type: accuracy value: 71.06 name: accuracy
• —task: type: text-generation name: OpenBookQA Facts dataset: name: OpenBookQA type: openbookqa split: test args: numfewshot: 5 metrics:
• —type: accuracy value: 30.60 name: accuracy
• —task: type: text-generation name: MMLU Knowledge Test dataset: name: MMLU type: mmlu config: all split: test args: numfewshot: 5 metrics:
• —type: accuracy value: 38.03 name: accuracy
• —task: type: text-generation name: Hellaswag Contextual Completions dataset: name: Hellaswag type: hellaswag split: validation args: numfewshot: 10 metrics:
• —type: accuracy value: 43.44 name: accuracy
• —task: type: text-generation name: GSM8k Mathematical Reasoning dataset: name: GSM8k type: gsm8k split: test args: numfewshot: 5 metrics:
• —type: accuracy value: 14.71 name: exact match (strict)
• —type: accuracy value: 14.63 name: exact match (flexible)
• —task: type: text-generation name: BoolQ Question Answering dataset: name: BoolQ type: boolq split: validation args: numfewshot: 5 metrics:
• —type: accuracy value: 65.20 name: accuracy
• —task: type: text-generation name: ARC Challenge dataset: name: ARC Challenge type: ai2arc split: test args: numfew_shot: 25 metrics:
• —type: accuracy value: 34.81 name: accuracy ---

🤏 Coven Tiny 1.1B 32K ORPO

Coven Tiny 1.1B 32K is an improved iteration of TinyLlama-1.1B-Chat-v1.0, refined to expand processing capabilities and refine language model preferences. This model includes a significantly increased context limit of 32K tokens, allowing for more extensive data processing and understanding of complex language scenarios. In addition, Coven Tiny 1.1B 32K uses the innovative ORPO (Monolithic Preference Optimization without Reference Model) technique. ORPO simplifies the fine-tuning process by directly optimizing the odds ratio to distinguish between favorable and unfavorable generation styles, effectively improving model performance without the need for an additional preference alignment step.

Model Details

• —Model name: Coven Tiny 1.1B 32K ORPO alpha
• —Fine-tuned by: raidhon
• —Base model: TinyLlama-1.1B-Chat-v1.0
• —Parameters: 1.1B
• —Context: 32K
• —Language(s): Multilingual
• —License: Apache2.0

Eval

💻 Usage

# Install transformers from source - only needed for versions <= v4.34
# pip install git+https://github.com/huggingface/transformers.git
# pip install accelerate

import torch
from transformers import pipeline

pipe = pipeline("text-generation", model="raidhon/coven_tiny_1.1b_32k_orpo_alpha", torch_dtype=torch.bfloat16, device_map="auto")

messages = [
    {
        "role": "system",
        "content": "You are a friendly chatbot who always responds in the style of a pirate",
    },
    {"role": "user", "content": "How many helicopters can a human eat in one sitting?"},
]
prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipe(prompt, max_new_tokens=2048, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])