prithivMLmods/QwQ-4B-Instruct

QwQ-4B-Instruct-Model-Files

The QwQ-4B-Instruct is a lightweight and efficient fine-tuned language model for instruction-following tasks and reasoning. It is based on a quantized version of the Qwen2.5-7B model, optimized for inference speed and reduced memory consumption, while retaining robust capabilities for complex tasks.

File Name	Size	Description	Upload Status
.gitattributes	1.57 kB	Tracks files stored with Git LFS.	Uploaded
README.md	271 Bytes	Basic project documentation.	Updated
added_tokens.json	657 Bytes	Specifies additional tokens for the tokenizer.	Uploaded
config.json	1.26 kB	Detailed model configuration file.	Uploaded
generation_config.json	281 Bytes	Configuration for text generation settings.	Uploaded
merges.txt	1.82 MB	Byte pair encoding (BPE) merge rules for tokenizer.	Uploaded
model-00001-of-00002.safetensors	4.46 GB	Part 1 of the model weights in safetensors format.	Uploaded (LFS)
model-00002-of-00002.safetensors	1.09 GB	Part 2 of the model weights in safetensors format.	Uploaded (LFS)
model.safetensors.index.json	124 kB	Index file for safetensors model sharding.	Uploaded
special_tokens_map.json	644 Bytes	Mapping of special tokens (e.g., , ).	Uploaded
tokenizer.json	11.4 MB	Complete tokenizer configuration.	Uploaded (LFS)
tokenizer_config.json	7.73 kB	Settings for the tokenizer integration.	Uploaded
vocab.json	2.78 MB	Vocabulary file containing token-to-id mappings.	Uploaded

Key Features:

1. Model Size:

   • 4.46B parameters.

2. Precision Support:

   • Available in multiple tensor types:
     • FP16
     • F32
     • U8 (Quantized)

3. Model Sharding:

   • The model weights are stored in two parts for efficient download:
     • model-00001-of-00002.safetensors (4.46 GB)
     • model-00002-of-00002.safetensors (1.09 GB)
   • Indexed with model.safetensors.index.json.

4. Tokenizer:

   • Uses Byte-Pair Encoding (BPE).
   • Includes:
     • vocab.json (2.78 MB)
     • merges.txt (1.82 MB)
     • tokenizer.json (11.4 MB, pre-trained configuration).
   • Special tokens mapped in special_tokens_map.json (e.g., <pad>, <eos>).

5. Configuration Files:

   • config.json: Defines the architecture, hyperparameters, and settings.
   • generation_config.json: Specifies text generation behavior (e.g., max length, temperature).

---

Training Dataset:

• Dataset Name: amphora/QwQ-LongCoT-130K
• Size: 133k examples.
• Focus: Chain-of-Thought reasoning for detailed and logical outputs.

---

Use Cases:

1. Instruction-Following:

   • Excels in handling concise and multi-step instructions.

2. Reasoning:

   • Well-suited for tasks requiring logical deductions and detailed explanations.

3. Text Generation:

   • Generates coherent and contextually aware responses across various domains.

4. Resource-Constrained Applications:

   • Optimized for scenarios requiring lower computational resources due to its smaller model size and quantization.

---