README.md

metadata

license: creativeml-openrail-m
datasets:
  - amphora/QwQ-LongCoT-130K
language:
  - en
base_model:
  - Qwen/Qwen2.5-7B-Instruct
pipeline_tag: text-generation
library_name: transformers
tags:
  - Long-CoT
  - Qwen2.5
  - 7B
  - safetensors
  - text-generation-inference
  - QwQ
  - FP16 precision
  - SFT
  - Math

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.

File Name	Size	Description	Upload Status
.gitattributes	1.57 kB	Tracks large files with Git LFS.	Uploaded
README.md	273 Bytes	Contains initial documentation, likely minimal.	Updated
added_tokens.json	657 Bytes	Maps additional tokens for the tokenizer.	Uploaded
config.json	848 Bytes	Model configuration (basic setup).	Uploaded
generation_config.json	281 Bytes	Settings for text generation tasks.	Uploaded
merges.txt	1.82 MB	Tokenizer merges for byte-pair encoding (BPE).	Uploaded
model-00001-of-00004.safetensors	4.88 GB	First part of model weights (split for LFS).	Uploaded (LFS)
model-00002-of-00004.safetensors	4.93 GB	Second part of model weights.	Uploaded (LFS)
model-00003-of-00004.safetensors	4.33 GB	Third part of model weights.	Uploaded (LFS)
model-00004-of-00004.safetensors	1.09 GB	Fourth part of model weights.	Uploaded (LFS)
model.safetensors.index.json	29.5 kB	Index file for managing model shards.	Uploaded
special_tokens_map.json	644 Bytes	Maps special tokens like <pad> or <eos>.	Uploaded
tokenizer.json	11.4 MB	Pre-trained tokenizer file in JSON format.	Uploaded (LFS)
tokenizer_config.json	7.73 kB	Configuration details for the tokenizer.	Uploaded
vocab.json	2.78 MB	Tokenizer vocabulary.	Uploaded

Sample Long CoT:

---

Key Features:

1. Model Size:

   • 7.62B parameters (FP16 precision).

2. Model Sharding:

   • The model weights are split into 4 shards (safetensors) for efficient storage and download:
     • model-00001-of-00004.safetensors (4.88 GB)
     • model-00002-of-00004.safetensors (4.93 GB)
     • model-00003-of-00004.safetensors (4.33 GB)
     • model-00004-of-00004.safetensors (1.09 GB)

3. Tokenizer:

   • Byte-pair encoding (BPE) based.
   • Files included:
     • vocab.json (2.78 MB)
     • merges.txt (1.82 MB)
     • tokenizer.json (11.4 MB)
   • Special tokens mapped in special_tokens_map.json (e.g., <pad>, <eos>).

4. Configuration Files:

   • config.json: Defines model architecture and hyperparameters.
   • 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. Instruction Following:
   Handle user instructions effectively, even for multi-step tasks.

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

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

---