prithivMLmods/QwQ-LCoT-7B-Instruct

QwQ-LCoT-7B-Instruct Model File

The QwQ-LCoT-7B-Instruct is a fine-tuned language model designed for advanced reasoning and instruction-following tasks. It leverages the Qwen2.5-7B base model and has been fine-tuned on the amphora/QwQ-LongCoT-130K dataset, focusing on chain-of-thought (CoT) reasoning. This model is optimized for tasks requiring logical reasoning, detailed explanations, and multi-step problem-solving, making it ideal for applications such as instruction-following, text generation, and complex reasoning tasks.

Quickstart with Transformers

Here provides a code snippet with apply_chat_template to show you how to load the tokenizer and model and how to generate contents.

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "prithivMLmods/QwQ-LCoT-7B-Instruct"

model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)

prompt = "How many r in strawberry."
messages = [
    {"role": "system", "content": "You are a helpful and harmless assistant. You are Qwen developed by Alibaba. You should think 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]

Sample Long CoT:

Key Features:

1. 1.Model Size:
2. 2.7.62B parameters (FP16 precision).

1. 1.Model Sharding:
2. 2.The model weights are split into 4 shards (safetensors) for efficient storage and download:
3. 3.model-00001-of-00004.safetensors (4.88 GB)
4. 4.model-00002-of-00004.safetensors (4.93 GB)
5. 5.model-00003-of-00004.safetensors (4.33 GB)
6. 6.model-00004-of-00004.safetensors (1.09 GB)

1. 1.Tokenizer:
2. 2.Byte-pair encoding (BPE) based.
3. 3.Files included:
4. 4.vocab.json (2.78 MB)
5. 5.merges.txt (1.82 MB)
6. 6.tokenizer.json (11.4 MB)
7. 7.Special tokens mapped in special_tokens_map.json (e.g., <pad>, <eos>).

1. 1.Configuration Files:
2. 2.config.json: Defines model architecture and hyperparameters.
3. 3.generation_config.json: Settings for inference and text generation tasks.

Training Dataset:

• —Dataset Name: amphora/QwQ-LongCoT-130K
• —Size: 133k examples.
• —Focus: Chain-of-Thought reasoning for complex tasks.

Use Cases:

1. 1.Instruction Following: Handle user instructions effectively, even for multi-step tasks.

1. 1.Reasoning Tasks: Perform logical reasoning and generate detailed step-by-step solutions.

1. 1.Text Generation: Generate coherent, context-aware responses. ---