52 lines
850 B
JSON
52 lines
850 B
JSON
{
|
|
"alpha_pattern": {},
|
|
"auto_mapping": null,
|
|
"base_model_name_or_path": "Qwen/Qwen3-4B",
|
|
"bias": "none",
|
|
"fan_in_fan_out": false,
|
|
"inference_mode": true,
|
|
"init_lora_weights": true,
|
|
"layer_replication": null,
|
|
"layers_pattern": "layers",
|
|
"layers_to_transform": [
|
|
20,
|
|
21,
|
|
22,
|
|
23,
|
|
24,
|
|
25,
|
|
26,
|
|
27,
|
|
28,
|
|
29,
|
|
30,
|
|
31,
|
|
32,
|
|
33,
|
|
34,
|
|
35
|
|
],
|
|
"loftq_config": {},
|
|
"lora_alpha": 160.0,
|
|
"lora_dropout": 0.0,
|
|
"megatron_config": null,
|
|
"megatron_core": "megatron.core",
|
|
"modules_to_save": null,
|
|
"peft_type": "LORA",
|
|
"r": 8,
|
|
"rank_pattern": {},
|
|
"revision": null,
|
|
"target_modules": [
|
|
"down_proj",
|
|
"gate_proj",
|
|
"k_proj",
|
|
"o_proj",
|
|
"q_proj",
|
|
"up_proj",
|
|
"v_proj"
|
|
],
|
|
"task_type": "CAUSAL_LM",
|
|
"use_dora": false,
|
|
"use_rslora": false
|
|
}
|