Instructions to use FerrellSyntheticIntelligence/fsi-anomaly with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use FerrellSyntheticIntelligence/fsi-anomaly with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf FerrellSyntheticIntelligence/fsi-anomaly # Run inference directly in the terminal: llama cli -hf FerrellSyntheticIntelligence/fsi-anomaly
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf FerrellSyntheticIntelligence/fsi-anomaly # Run inference directly in the terminal: llama cli -hf FerrellSyntheticIntelligence/fsi-anomaly
Use pre-built binary
# Download pre-built binary from: # https://github.com/ggerganov/llama.cpp/releases # Start a local OpenAI-compatible server with a web UI: ./llama-server -hf FerrellSyntheticIntelligence/fsi-anomaly # Run inference directly in the terminal: ./llama-cli -hf FerrellSyntheticIntelligence/fsi-anomaly
Build from source code
git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-server llama-cli # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama-server -hf FerrellSyntheticIntelligence/fsi-anomaly # Run inference directly in the terminal: ./build/bin/llama-cli -hf FerrellSyntheticIntelligence/fsi-anomaly
Use Docker
docker model run hf.co/FerrellSyntheticIntelligence/fsi-anomaly
- LM Studio
- Jan
- Ollama
How to use FerrellSyntheticIntelligence/fsi-anomaly with Ollama:
ollama run hf.co/FerrellSyntheticIntelligence/fsi-anomaly
- Unsloth Desktop
- Docker Model Runner
How to use FerrellSyntheticIntelligence/fsi-anomaly with Docker Model Runner:
docker model run hf.co/FerrellSyntheticIntelligence/fsi-anomaly
- Lemonade
How to use FerrellSyntheticIntelligence/fsi-anomaly with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull FerrellSyntheticIntelligence/fsi-anomaly
Run and chat with the model
lemonade run user.fsi-anomaly-{{QUANT_TAG}}List all available models
lemonade list
- Atomic Chat
| """Unit tests for train/ties_merge.py and research/rlvr.py. | |
| Run: .venv/bin/python tests/test_posttrain.py | |
| """ | |
| import sys | |
| from pathlib import Path | |
| sys.path.insert(0, str(Path(__file__).resolve().parents[1])) | |
| import torch | |
| from train.ties_merge import ties_merge, trim_delta | |
| from research.rlvr import reward, reward_card | |
| def test_trim_keeps_topk(): | |
| d = torch.tensor([1.0, -0.5, 0.01, 0.001, 0.0005]) | |
| t = trim_delta(d, keep=0.4) | |
| assert t[0] == 1.0 and t[1] == -0.5 | |
| assert t[2] == 0.0 and t[3] == 0.0 | |
| def test_ties_sign_consensus(): | |
| base = {"w": torch.zeros(4)} | |
| t1 = {"w": torch.tensor([1.0, 1.0, 1.0, -1.0])} | |
| t2 = {"w": torch.tensor([1.0, -1.0, 1.0, -1.0])} | |
| out = ties_merge(base, [t1, t2], keep=1.0) | |
| # sign agreement at idx 0, 2, 3 -> merge; idx 1 disagrees -> zero | |
| assert out["w"][0] == 1.0 | |
| assert out["w"][2] == 1.0 | |
| assert out["w"][3] == -1.0 | |
| assert out["w"][1] == 0.0 | |
| def test_rlvr_reward_correct_with_citation(): | |
| r = reward(gold="refutes", policy="refutes", citation="1982", | |
| evidence="the deed file states 1982") | |
| assert r["verdict"] == 1.0 and r["citation"] == 0.2 and r["total"] == 1.2 | |
| def test_rlvr_reward_wrong_verdict(): | |
| r = reward(gold="refutes", policy="supports", citation="1982", | |
| evidence="the deed file states 1982") | |
| assert r["verdict"] == -1.0 and r["total"] == -0.8 | |
| def test_rlvr_reward_abstain_is_zero(): | |
| r = reward(gold="refutes", policy="not enough information", | |
| citation="", evidence="the deed file states 1982") | |
| assert r["verdict"] == 0.0 and r["total"] == 0.0 | |
| def test_rlvr_reward_false_citation_penalty(): | |
| r = reward(gold="supports", policy="supports", citation="1978", | |
| evidence="the deed file states 1982") | |
| assert r["verdict"] == 1.0 and r["citation"] == -0.2 and r["total"] == 0.8 | |
| def test_rlvr_card_trace(): | |
| card = reward_card(gold="refutes", policy="refutes", citation="1982", | |
| evidence="the deed file states 1982", probe="rt05") | |
| assert card["probe"] == "rt05" and card["total"] == 1.2 | |
| for k in ("gold", "policy", "verdict", "citation", "total"): | |
| assert k in card | |
| if __name__ == "__main__": | |
| fns = [v for k, v in sorted(globals().items()) if k.startswith("test_")] | |
| for fn in fns: | |
| fn() | |
| print(f"ok {fn.__name__}") | |
| print(f"\n{len(fns)} posttrain tests passed") | |