Text Classification
MLX
Safetensors
English
decision-model
kev
crypto
news-classification
trading
lora
qwen3.5
apple-silicon
novaeon
Instructions to use NovaeonStudio/novaeon-sentinel-9b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use NovaeonStudio/novaeon-sentinel-9b with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download NovaeonStudio/novaeon-sentinel-9b --local-dir novaeon-sentinel-9b
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
File size: 1,473 Bytes
4ea0acf | 1 2 3 4 5 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 | {
"args": {
"base": "Qwen/Qwen3.5-9B-Base",
"n_per_source": 1000,
"epochs": 2,
"lr": 3e-05,
"head_lr": 0.0,
"weight_decay": 0.01,
"lora": 16,
"accum": 8,
"holdout": "",
"perm_kl": 0.0,
"perm_frac": 0.3,
"ord_w": 0.0,
"label_smoothing": 0.0,
"brier_w": 0.0,
"focal_gamma": 0.0,
"suite": null,
"train_sources": "",
"device": "mps",
"batch": 1,
"dtype": "fp32",
"weights_dtype": "bf16",
"checkpointing": 0,
"option_isolation": 0,
"special_embeddings": 0,
"head_dim": 256,
"lora_targets": "all",
"base_revision": "",
"p_none": 0.1,
"p_none_distract": 0.12,
"p_distract": 0.15,
"p_none_pair": 0.0,
"synthetic_repeat": 1,
"public_frac": 1.0,
"anchor": "",
"anchor_w": 0.0,
"anchor_sources": "",
"out": "runs/novaeon-sentinel-9b",
"data": "data/train-v3-2q.jsonl",
"max_state": 704,
"replay": 0,
"init_from": "jaredpalmer/kev-9b",
"seed": 1
},
"suite_sha256": null,
"base_revision": null,
"init_source": {
"init_from": "jaredpalmer/kev-9b",
"resolved": "jaredpalmer/kev-9b@2629c06a5aeb0feb3b9783bafed17ed8f39ecf5c",
"adapter_sha256": "c02d0d1a70367b5779deb23b28153284b59c067038323cc17ddab5b92dfce72f",
"head_sha256": "5528f777437f561cb4a9d56b0278a42b6ae6edd373e7921e52b13adaac988c9f",
"adapter_tensors": 496
},
"ordinal_objective": "ranked_probability_score",
"holdout": []
}
|