Upload LoRA adapter (README written by author)
Browse files- README.md +2 -2
- adapter_config.json +5 -17
- adapter_model.safetensors +1 -1
README.md
CHANGED
|
@@ -34,8 +34,8 @@ while intermediate reasoning (Chain-of-Thought) is masked.
|
|
| 34 |
- Base model: Qwen/Qwen3-4B-Instruct-2507
|
| 35 |
- Method: QLoRA (4-bit)
|
| 36 |
- Max sequence length: 512
|
| 37 |
-
- Epochs:
|
| 38 |
-
- Learning rate: 1e-
|
| 39 |
- LoRA: r=64, alpha=128
|
| 40 |
|
| 41 |
## Usage
|
|
|
|
| 34 |
- Base model: Qwen/Qwen3-4B-Instruct-2507
|
| 35 |
- Method: QLoRA (4-bit)
|
| 36 |
- Max sequence length: 512
|
| 37 |
+
- Epochs: 1
|
| 38 |
+
- Learning rate: 1e-05
|
| 39 |
- LoRA: r=64, alpha=128
|
| 40 |
|
| 41 |
## Usage
|
adapter_config.json
CHANGED
|
@@ -1,7 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"alora_invocation_tokens": null,
|
| 3 |
"alpha_pattern": {},
|
| 4 |
-
"arrow_config": null,
|
| 5 |
"auto_mapping": {
|
| 6 |
"base_model_class": "Qwen3ForCausalLM",
|
| 7 |
"parent_library": "transformers.models.qwen3.modeling_qwen3",
|
|
@@ -9,10 +7,6 @@
|
|
| 9 |
},
|
| 10 |
"base_model_name_or_path": "unsloth/qwen3-4b-instruct-2507-unsloth-bnb-4bit",
|
| 11 |
"bias": "none",
|
| 12 |
-
"corda_config": null,
|
| 13 |
-
"ensure_weight_tying": false,
|
| 14 |
-
"eva_config": null,
|
| 15 |
-
"exclude_modules": null,
|
| 16 |
"fan_in_fan_out": false,
|
| 17 |
"inference_mode": true,
|
| 18 |
"init_lora_weights": true,
|
|
@@ -21,30 +15,24 @@
|
|
| 21 |
"layers_to_transform": null,
|
| 22 |
"loftq_config": {},
|
| 23 |
"lora_alpha": 128,
|
| 24 |
-
"lora_bias": false,
|
| 25 |
"lora_dropout": 0.0,
|
| 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": 64,
|
| 33 |
"rank_pattern": {},
|
| 34 |
"revision": null,
|
| 35 |
"target_modules": [
|
| 36 |
-
"gate_proj",
|
| 37 |
-
"down_proj",
|
| 38 |
"k_proj",
|
| 39 |
-
"v_proj",
|
| 40 |
-
"q_proj",
|
| 41 |
"up_proj",
|
| 42 |
-
"o_proj"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 43 |
],
|
| 44 |
-
"target_parameters": null,
|
| 45 |
"task_type": "CAUSAL_LM",
|
| 46 |
-
"trainable_token_indices": null,
|
| 47 |
"use_dora": false,
|
| 48 |
-
"use_qalora": false,
|
| 49 |
"use_rslora": false
|
| 50 |
}
|
|
|
|
| 1 |
{
|
|
|
|
| 2 |
"alpha_pattern": {},
|
|
|
|
| 3 |
"auto_mapping": {
|
| 4 |
"base_model_class": "Qwen3ForCausalLM",
|
| 5 |
"parent_library": "transformers.models.qwen3.modeling_qwen3",
|
|
|
|
| 7 |
},
|
| 8 |
"base_model_name_or_path": "unsloth/qwen3-4b-instruct-2507-unsloth-bnb-4bit",
|
| 9 |
"bias": "none",
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
"fan_in_fan_out": false,
|
| 11 |
"inference_mode": true,
|
| 12 |
"init_lora_weights": true,
|
|
|
|
| 15 |
"layers_to_transform": null,
|
| 16 |
"loftq_config": {},
|
| 17 |
"lora_alpha": 128,
|
|
|
|
| 18 |
"lora_dropout": 0.0,
|
| 19 |
"megatron_config": null,
|
| 20 |
"megatron_core": "megatron.core",
|
| 21 |
"modules_to_save": null,
|
| 22 |
"peft_type": "LORA",
|
|
|
|
|
|
|
| 23 |
"r": 64,
|
| 24 |
"rank_pattern": {},
|
| 25 |
"revision": null,
|
| 26 |
"target_modules": [
|
|
|
|
|
|
|
| 27 |
"k_proj",
|
|
|
|
|
|
|
| 28 |
"up_proj",
|
| 29 |
+
"o_proj",
|
| 30 |
+
"down_proj",
|
| 31 |
+
"gate_proj",
|
| 32 |
+
"v_proj",
|
| 33 |
+
"q_proj"
|
| 34 |
],
|
|
|
|
| 35 |
"task_type": "CAUSAL_LM",
|
|
|
|
| 36 |
"use_dora": false,
|
|
|
|
| 37 |
"use_rslora": false
|
| 38 |
}
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 528550256
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6768be3e28c690f9dec1ad4119e341132b0af196bb3a61b424e927a24fe3452
|
| 3 |
size 528550256
|