Upload adapter_config.json with huggingface_hub
Browse files- adapter_config.json +40 -0
adapter_config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"adapter_path": "/Volumes/Virtual Server/projects/psm-ops/multichain-day-model/model-output/mlx-adapters",
|
| 3 |
+
"batch_size": 1,
|
| 4 |
+
"config": null,
|
| 5 |
+
"data": "/Volumes/Virtual Server/projects/psm-ops/multichain-day-model/model-output/mlx-training-data",
|
| 6 |
+
"fine_tune_type": "lora",
|
| 7 |
+
"grad_accumulation_steps": 1,
|
| 8 |
+
"grad_checkpoint": true,
|
| 9 |
+
"iters": 200,
|
| 10 |
+
"learning_rate": 1e-05,
|
| 11 |
+
"lora_parameters": {
|
| 12 |
+
"rank": 8,
|
| 13 |
+
"dropout": 0.0,
|
| 14 |
+
"scale": 20.0
|
| 15 |
+
},
|
| 16 |
+
"lr_schedule": null,
|
| 17 |
+
"mask_prompt": false,
|
| 18 |
+
"max_seq_length": 1024,
|
| 19 |
+
"model": "mlx-community/DeepSeek-R1-Distill-Llama-8B-4bit",
|
| 20 |
+
"num_layers": 4,
|
| 21 |
+
"optimizer": "adam",
|
| 22 |
+
"optimizer_config": {
|
| 23 |
+
"adam": {},
|
| 24 |
+
"adamw": {},
|
| 25 |
+
"muon": {},
|
| 26 |
+
"sgd": {},
|
| 27 |
+
"adafactor": {}
|
| 28 |
+
},
|
| 29 |
+
"project_name": null,
|
| 30 |
+
"report_to": null,
|
| 31 |
+
"resume_adapter_file": null,
|
| 32 |
+
"save_every": 100,
|
| 33 |
+
"seed": 42,
|
| 34 |
+
"steps_per_eval": 50,
|
| 35 |
+
"steps_per_report": 10,
|
| 36 |
+
"test": false,
|
| 37 |
+
"test_batches": 500,
|
| 38 |
+
"train": true,
|
| 39 |
+
"val_batches": 25
|
| 40 |
+
}
|