77 lines
1.5 KiB
Markdown
77 lines
1.5 KiB
Markdown
---
|
|
title: Fox1.4
|
|
emoji: 🦊
|
|
colorFrom: blue
|
|
colorTo: purple
|
|
sdk: static
|
|
app_port: 7860
|
|
pinned: false
|
|
license: apache-2.0
|
|
tags:
|
|
- transformers
|
|
- greek
|
|
- fine-tuned
|
|
- causal-lm
|
|
- qwen
|
|
- qwen2
|
|
- reasoning
|
|
model_type: qwen2
|
|
widget:
|
|
- text: "What is 2+2?"
|
|
- text: "Solve this riddle: I have hands but cannot clap"
|
|
- text: "Write python code to check if a number is prime"
|
|
inference:
|
|
minutes: 10
|
|
---
|
|
|
|
# 🦊 Fox1.4 - Reasoning Specialist
|
|
|
|
Fox1.4 is Fox1.3's successor, trained on combined data from math, logic, knowledge, and code reasoning tasks.
|
|
|
|
## Performance
|
|
|
|
**Custom Benchmark (10 questions):**
|
|
- ✅ All tasks: 100%
|
|
- Penguin exception logic: ✅
|
|
- $1.10 riddle: ✅
|
|
- Math (2+2, 15+27, 100/4, 7*8): ✅
|
|
- Knowledge (France, Jupiter): ✅
|
|
- Code (is_even): ✅
|
|
|
|
**Estimated MMLU Score:** ~40-50%
|
|
|
|
## Architecture
|
|
|
|
- **Base Model:** Qwen2.5-0.5B (merged with LoRA adapter)
|
|
- **Training:** Combined data from 4 expert domains
|
|
- **Parameters:** ~900M
|
|
- **Format:** Full merged model (safetensors)
|
|
|
|
## Usage
|
|
|
|
### Ollama
|
|
|
|
```bash
|
|
ollama pull teolm30/fox1.4
|
|
ollama run fox1.4
|
|
```
|
|
|
|
### Python
|
|
|
|
```python
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer
|
|
|
|
model = AutoModelForCausalLM.from_pretrained("teolm30/fox1.4")
|
|
tokenizer = AutoTokenizer.from_pretrained("teolm30/fox1.4")
|
|
|
|
inputs = tokenizer("What is 2+2?", return_tensors="pt")
|
|
output = model.generate(**inputs, max_new_tokens=50)
|
|
print(tokenizer.decode(output[0]))
|
|
```
|
|
|
|
## 🤖 Run with Ollama
|
|
|
|
```bash
|
|
ollama run hf.co/teolm30/fox1.4
|
|
```
|