Views
No views yet
| Parameter | Value |
|---|---|
| Base model | google/gemma-4-E4B-it (4B params) |
| Dataset size | 56,612 examples |
| Method | LoRA (8 layers) |
| Iterations | 500 |
| Learning rate | 1e-5 |
| Train loss | 0.120 |
| Validation loss | 0.153 |
| Training time | ~1.5 hours on Mac Studio M2 Ultra |
| Framework | mlx-lm |
| Source | Examples |
|---|---|
| Cogneo Crypto Sentiment | 43,027 |
| Forta Malicious Contracts | 12,000 |
| HuggingFace open datasets (Bitcoin, smart contracts, blockchain) | 3,267 |
| ChainGPT-generated Q&A (16 categories) | ~300 |
| Total (after dedup) | 56,612 |
1# Download and run
2ollama run cryptogemma
3
4# Example
5>>> Analyze Bitcoin's current market structure and key support levels1from transformers import AutoTokenizer, AutoModelForCausalLM
2
3model = AutoModelForCausalLM.from_pretrained("SkillForge/CryptoGemma-4B-v1")
4tokenizer = AutoTokenizer.from_pretrained("SkillForge/CryptoGemma-4B-v1")
5
6prompt = "Analyze the risk factors of this DeFi protocol..."
7inputs = tokenizer(prompt, return_tensors="pt")
8outputs = model.generate(**inputs, max_new_tokens=512)
9print(tokenizer.decode(outputs[0], skip_special_tokens=True))1from mlx_lm import load, generate
2
3model, tokenizer = load("SkillForge/CryptoGemma-4B-v1")
4response = generate(model, tokenizer, prompt="What are the key metrics for evaluating a DeFi token?", max_tokens=512)1@misc{cryptogemma2026,
2 title={CryptoGemma-4B-v1: Open-Source Crypto-Specialized LLM},
3 author={SkillForge Team},
4 year={2026},
5 url={https://huggingface.co/SkillForge/CryptoGemma-4B-v1}
6}