Upload LoRA adapter (README written by author)
Browse files- README.md +3 -3
- adapter_config.json +9 -6
- adapter_model.safetensors +2 -2
README.md
CHANGED
|
@@ -13,7 +13,7 @@ tags:
|
|
| 13 |
- structured-output
|
| 14 |
---
|
| 15 |
|
| 16 |
-
qwen3-4b-structured-output-
|
| 17 |
|
| 18 |
This repository provides a **LoRA adapter** fine-tuned from
|
| 19 |
**Qwen/Qwen3-4B-Instruct-2507** using **QLoRA (4-bit, Unsloth)**.
|
|
@@ -35,8 +35,8 @@ while intermediate reasoning (Chain-of-Thought) is masked.
|
|
| 35 |
- Method: QLoRA (4-bit)
|
| 36 |
- Max sequence length: 512
|
| 37 |
- Epochs: 2
|
| 38 |
-
- Learning rate: 1e-
|
| 39 |
-
- LoRA: r=
|
| 40 |
|
| 41 |
## Usage
|
| 42 |
|
|
|
|
| 13 |
- structured-output
|
| 14 |
---
|
| 15 |
|
| 16 |
+
qwen3-4b-structured-output-lora_4th
|
| 17 |
|
| 18 |
This repository provides a **LoRA adapter** fine-tuned from
|
| 19 |
**Qwen/Qwen3-4B-Instruct-2507** using **QLoRA (4-bit, Unsloth)**.
|
|
|
|
| 35 |
- Method: QLoRA (4-bit)
|
| 36 |
- Max sequence length: 512
|
| 37 |
- Epochs: 2
|
| 38 |
+
- Learning rate: 1e-06
|
| 39 |
+
- LoRA: r=64, alpha=128
|
| 40 |
|
| 41 |
## Usage
|
| 42 |
|
adapter_config.json
CHANGED
|
@@ -20,23 +20,26 @@
|
|
| 20 |
"layers_pattern": null,
|
| 21 |
"layers_to_transform": null,
|
| 22 |
"loftq_config": {},
|
| 23 |
-
"lora_alpha":
|
| 24 |
"lora_bias": false,
|
| 25 |
-
"lora_dropout": 0.
|
| 26 |
"megatron_config": null,
|
| 27 |
"megatron_core": "megatron.core",
|
| 28 |
"modules_to_save": null,
|
| 29 |
"peft_type": "LORA",
|
| 30 |
"peft_version": "0.18.1",
|
| 31 |
"qalora_group_size": 16,
|
| 32 |
-
"r":
|
| 33 |
"rank_pattern": {},
|
| 34 |
"revision": null,
|
| 35 |
"target_modules": [
|
| 36 |
-
"
|
| 37 |
-
"",
|
| 38 |
"q_proj",
|
| 39 |
-
"v_proj"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
],
|
| 41 |
"target_parameters": null,
|
| 42 |
"task_type": "CAUSAL_LM",
|
|
|
|
| 20 |
"layers_pattern": null,
|
| 21 |
"layers_to_transform": null,
|
| 22 |
"loftq_config": {},
|
| 23 |
+
"lora_alpha": 128,
|
| 24 |
"lora_bias": false,
|
| 25 |
+
"lora_dropout": 0.05,
|
| 26 |
"megatron_config": null,
|
| 27 |
"megatron_core": "megatron.core",
|
| 28 |
"modules_to_save": null,
|
| 29 |
"peft_type": "LORA",
|
| 30 |
"peft_version": "0.18.1",
|
| 31 |
"qalora_group_size": 16,
|
| 32 |
+
"r": 64,
|
| 33 |
"rank_pattern": {},
|
| 34 |
"revision": null,
|
| 35 |
"target_modules": [
|
| 36 |
+
"gate_proj",
|
|
|
|
| 37 |
"q_proj",
|
| 38 |
+
"v_proj",
|
| 39 |
+
"down_proj",
|
| 40 |
+
"up_proj",
|
| 41 |
+
"o_proj",
|
| 42 |
+
"k_proj"
|
| 43 |
],
|
| 44 |
"target_parameters": null,
|
| 45 |
"task_type": "CAUSAL_LM",
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d9f6f6596487e64a7542979ce0cc273a38736fd212a9a90b78106106a15f1ee6
|
| 3 |
+
size 528550256
|