Update model after training on balanced dataset (95.6K examples, 3 epochs)
Browse files- adapter_config.json +12 -5
adapter_config.json
CHANGED
|
@@ -6,12 +6,17 @@
|
|
| 6 |
"fan_in_fan_out": false,
|
| 7 |
"inference_mode": true,
|
| 8 |
"init_lora_weights": true,
|
|
|
|
| 9 |
"layers_pattern": null,
|
| 10 |
"layers_to_transform": null,
|
| 11 |
"loftq_config": {},
|
| 12 |
"lora_alpha": 32,
|
| 13 |
"lora_dropout": 0.05,
|
|
|
|
|
|
|
| 14 |
"modules_to_save": [
|
|
|
|
|
|
|
| 15 |
"score"
|
| 16 |
],
|
| 17 |
"peft_type": "LORA",
|
|
@@ -19,13 +24,15 @@
|
|
| 19 |
"rank_pattern": {},
|
| 20 |
"revision": null,
|
| 21 |
"target_modules": [
|
| 22 |
-
"
|
|
|
|
| 23 |
"down_proj",
|
| 24 |
"k_proj",
|
| 25 |
-
"
|
| 26 |
"gate_proj",
|
| 27 |
-
"up_proj",
|
| 28 |
"o_proj"
|
| 29 |
],
|
| 30 |
-
"task_type": "SEQ_CLS"
|
| 31 |
-
|
|
|
|
|
|
|
|
|
| 6 |
"fan_in_fan_out": false,
|
| 7 |
"inference_mode": true,
|
| 8 |
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
"layers_pattern": null,
|
| 11 |
"layers_to_transform": null,
|
| 12 |
"loftq_config": {},
|
| 13 |
"lora_alpha": 32,
|
| 14 |
"lora_dropout": 0.05,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
"modules_to_save": [
|
| 18 |
+
"score",
|
| 19 |
+
"classifier",
|
| 20 |
"score"
|
| 21 |
],
|
| 22 |
"peft_type": "LORA",
|
|
|
|
| 24 |
"rank_pattern": {},
|
| 25 |
"revision": null,
|
| 26 |
"target_modules": [
|
| 27 |
+
"up_proj",
|
| 28 |
+
"v_proj",
|
| 29 |
"down_proj",
|
| 30 |
"k_proj",
|
| 31 |
+
"q_proj",
|
| 32 |
"gate_proj",
|
|
|
|
| 33 |
"o_proj"
|
| 34 |
],
|
| 35 |
+
"task_type": "SEQ_CLS",
|
| 36 |
+
"use_dora": false,
|
| 37 |
+
"use_rslora": false
|
| 38 |
+
}
|