Improve language tag
#2
by
lbourdois
- opened
README.md
CHANGED
|
@@ -1,108 +1,120 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: creativeml-openrail-m
|
| 3 |
-
datasets:
|
| 4 |
-
- amphora/QwQ-LongCoT-130K
|
| 5 |
-
language:
|
| 6 |
-
-
|
| 7 |
-
|
| 8 |
-
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
-
|
| 13 |
-
-
|
| 14 |
-
-
|
| 15 |
-
-
|
| 16 |
-
-
|
| 17 |
-
-
|
| 18 |
-
-
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
|
| 37 |
-
|
| 38 |
-
| `
|
| 39 |
-
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
|
| 53 |
-
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
-
|
| 60 |
-
|
| 61 |
-
|
| 62 |
-
|
| 63 |
-
|
| 64 |
-
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
|
| 68 |
-
|
| 69 |
-
|
| 70 |
-
|
| 71 |
-
|
| 72 |
-
|
| 73 |
-
|
| 74 |
-
|
| 75 |
-
|
| 76 |
-
|
| 77 |
-
|
| 78 |
-
|
| 79 |
-
|
| 80 |
-
|
| 81 |
-
|
| 82 |
-
|
| 83 |
-
|
| 84 |
-
|
| 85 |
-
|
| 86 |
-
|
| 87 |
-
|
| 88 |
-
|
| 89 |
-
|
| 90 |
-
|
| 91 |
-
|
| 92 |
-
|
| 93 |
-
|
| 94 |
-
|
| 95 |
-
|
| 96 |
-
|
| 97 |
-
|
| 98 |
-
|
| 99 |
-
|
| 100 |
-
|
| 101 |
-
|
| 102 |
-
|
| 103 |
-
|
| 104 |
-
|
| 105 |
-
|
| 106 |
-
|
| 107 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 108 |
---
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: creativeml-openrail-m
|
| 3 |
+
datasets:
|
| 4 |
+
- amphora/QwQ-LongCoT-130K
|
| 5 |
+
language:
|
| 6 |
+
- zho
|
| 7 |
+
- eng
|
| 8 |
+
- fra
|
| 9 |
+
- spa
|
| 10 |
+
- por
|
| 11 |
+
- deu
|
| 12 |
+
- ita
|
| 13 |
+
- rus
|
| 14 |
+
- jpn
|
| 15 |
+
- kor
|
| 16 |
+
- vie
|
| 17 |
+
- tha
|
| 18 |
+
- ara
|
| 19 |
+
base_model:
|
| 20 |
+
- Qwen/Qwen2.5-3B-Instruct
|
| 21 |
+
pipeline_tag: text-generation
|
| 22 |
+
library_name: transformers
|
| 23 |
+
tags:
|
| 24 |
+
- text-generation-inference
|
| 25 |
+
- long-CoT
|
| 26 |
+
- safetensors
|
| 27 |
+
- 3B
|
| 28 |
+
- Instruct
|
| 29 |
+
- QwQ
|
| 30 |
+
- Qwen2.5
|
| 31 |
+
---
|
| 32 |
+
### **QwQ-LCoT-3B-Instruct Model Card**
|
| 33 |
+
|
| 34 |
+
The **QwQ-LCoT-3B-Instruct** model is a lightweight, instruction-tuned language model designed for complex reasoning and explanation tasks. It is fine-tuned on the **Qwen2.5-3B-Instruct** base model using the **QwQ-LongCoT-130K** dataset, focusing on **long-chain-of-thought (LCoT)** reasoning for enhanced logical comprehension and detailed output generation.
|
| 35 |
+
|
| 36 |
+
| **File Name** | **Size** | **Description** | **Upload Status** |
|
| 37 |
+
|----------------------------------------|----------------|-------------------------------------------------|--------------------|
|
| 38 |
+
| `.gitattributes` | 1.57 kB | Specifies LFS tracking for large files. | Uploaded |
|
| 39 |
+
| `README.md` | 267 Bytes | Basic project information file. | Updated |
|
| 40 |
+
| `added_tokens.json` | 657 Bytes | Custom tokens added to the tokenizer. | Uploaded |
|
| 41 |
+
| `config.json` | 859 Bytes | Configuration file for the model. | Uploaded |
|
| 42 |
+
| `generation_config.json` | 281 Bytes | Configuration file for text generation settings.| Uploaded |
|
| 43 |
+
| `merges.txt` | 1.82 MB | Contains the byte-pair encoding (BPE) merges. | Uploaded |
|
| 44 |
+
| `pytorch_model-00001-of-00002.bin` | 4.96 GB | First shard of the model weights in PyTorch format. | Uploaded (LFS) |
|
| 45 |
+
| `pytorch_model-00002-of-00002.bin` | 1.21 GB | Second shard of the model weights in PyTorch format. | Uploaded (LFS) |
|
| 46 |
+
| `pytorch_model.bin.index.json` | 36 kB | Index mapping for sharded model weights. | Uploaded |
|
| 47 |
+
| `special_tokens_map.json` | 644 Bytes | Maps special tokens to their roles. | Uploaded |
|
| 48 |
+
| `tokenizer.json` | 11.4 MB | Serialized tokenizer data. | Uploaded (LFS) |
|
| 49 |
+
| `tokenizer_config.json` | 7.73 kB | Tokenizer configuration settings. | Uploaded |
|
| 50 |
+
| `vocab.json` | 2.78 MB | Vocabulary file for the tokenizer. | Uploaded |
|
| 51 |
+
|
| 52 |
+
### **Sample Long CoT:**
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
### **Key Features:**
|
| 57 |
+
|
| 58 |
+
1. **Long Chain-of-Thought Reasoning:**
|
| 59 |
+
- Specifically designed to generate comprehensive, step-by-step explanations for complex queries.
|
| 60 |
+
|
| 61 |
+
2. **Lightweight and Efficient:**
|
| 62 |
+
- With only 3 billion parameters, it is optimized for systems with limited computational resources without compromising reasoning capabilities.
|
| 63 |
+
|
| 64 |
+
3. **Instruction Optimization:**
|
| 65 |
+
- Fine-tuned to follow prompts and provide concise, actionable, and structured responses.
|
| 66 |
+
|
| 67 |
+
---
|
| 68 |
+
|
| 69 |
+
### **Training Details:**
|
| 70 |
+
|
| 71 |
+
- **Base Model:** [Qwen2.5-3B-Instruct](#)
|
| 72 |
+
- **Dataset:** [amphora/QwQ-LongCoT-130K](#)
|
| 73 |
+
- Comprising 133,000 annotated samples focusing on logical tasks and structured thinking.
|
| 74 |
+
---
|
| 75 |
+
|
| 76 |
+
### **Capabilities:**
|
| 77 |
+
|
| 78 |
+
1. **Text Generation:**
|
| 79 |
+
- Provides detailed, structured, and logical text outputs tailored to user prompts.
|
| 80 |
+
|
| 81 |
+
2. **Reasoning Tasks:**
|
| 82 |
+
- Solves step-by-step problems in math, logic, and science.
|
| 83 |
+
|
| 84 |
+
3. **Educational Assistance:**
|
| 85 |
+
- Generates coherent explanations for academic and research purposes.
|
| 86 |
+
|
| 87 |
+
4. **Dialogue and Summarization:**
|
| 88 |
+
- Handles conversational queries and summarizes long documents effectively.
|
| 89 |
+
|
| 90 |
+
---
|
| 91 |
+
|
| 92 |
+
### **Usage Instructions:**
|
| 93 |
+
|
| 94 |
+
1. **Setup:**
|
| 95 |
+
Download all model files and ensure compatibility with the Hugging Face Transformers library.
|
| 96 |
+
|
| 97 |
+
2. **Loading the Model:**
|
| 98 |
+
```python
|
| 99 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 100 |
+
|
| 101 |
+
model_name = "prithivMLmods/QwQ-LCoT-3B-Instruct"
|
| 102 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
| 103 |
+
model = AutoModelForCausalLM.from_pretrained(model_name)
|
| 104 |
+
```
|
| 105 |
+
|
| 106 |
+
3. **Generate Long-Chain Reasoning Outputs:**
|
| 107 |
+
```python
|
| 108 |
+
input_text = "Explain the process of photosynthesis step-by-step."
|
| 109 |
+
inputs = tokenizer(input_text, return_tensors="pt")
|
| 110 |
+
outputs = model.generate(**inputs, max_length=300, temperature=0.5)
|
| 111 |
+
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
|
| 112 |
+
```
|
| 113 |
+
|
| 114 |
+
4. **Customize Output Generation:**
|
| 115 |
+
Modify the `generation_config.json` file for different scenarios:
|
| 116 |
+
- **`temperature`**: Controls randomness (lower = deterministic, higher = creative).
|
| 117 |
+
- **`max_length`**: Sets response length.
|
| 118 |
+
- **`top_p`**: Adjusts sampling for diversity in outputs.
|
| 119 |
+
|
| 120 |
---
|