41 lines
1.1 KiB
JSON
41 lines
1.1 KiB
JSON
{
|
|
"adapter_path": "/Users/amitashwini/Projects/echo-flow-smart-dictation/ModelTraining/data/models/mumble-cleanup-v2-mlx/adapters",
|
|
"batch_size": 4,
|
|
"clear_cache_threshold": 0,
|
|
"config": null,
|
|
"data": "/Users/amitashwini/Projects/echo-flow-smart-dictation/ModelTraining/data/mlx_dataset",
|
|
"fine_tune_type": "lora",
|
|
"grad_accumulation_steps": 4,
|
|
"grad_checkpoint": false,
|
|
"iters": 2000,
|
|
"learning_rate": 0.0002,
|
|
"lora_parameters": {
|
|
"rank": 8,
|
|
"dropout": 0.0,
|
|
"scale": 20.0
|
|
},
|
|
"lr_schedule": null,
|
|
"mask_prompt": true,
|
|
"max_seq_length": 1024,
|
|
"model": "Qwen/Qwen2.5-0.5B-Instruct",
|
|
"num_layers": 16,
|
|
"optimizer": "adam",
|
|
"optimizer_config": {
|
|
"adam": {},
|
|
"adamw": {},
|
|
"muon": {},
|
|
"sgd": {},
|
|
"adafactor": {}
|
|
},
|
|
"project_name": null,
|
|
"report_to": null,
|
|
"resume_adapter_file": null,
|
|
"save_every": 200,
|
|
"seed": 42,
|
|
"steps_per_eval": 100,
|
|
"steps_per_report": 20,
|
|
"test": false,
|
|
"test_batches": 500,
|
|
"train": true,
|
|
"val_batches": 25
|
|
} |