Daniel Doña
commited on
Commit
·
36ae20c
1
Parent(s):
0afe5ee
Initial test
Browse files- README.md +25 -0
- adapter_config.json +32 -0
- adapter_model.safetensors +3 -0
README.md
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model:
|
| 3 |
+
- Qwen/Qwen3-30B-A3B-Instruct-2507
|
| 4 |
+
- Qwen/Qwen3-Coder-30B-A3B-Instruct
|
| 5 |
+
library_name: peft
|
| 6 |
+
tags:
|
| 7 |
+
- mergekit
|
| 8 |
+
- peft
|
| 9 |
+
|
| 10 |
+
---
|
| 11 |
+
# output_path
|
| 12 |
+
|
| 13 |
+
This is a LoRA extracted from a language model. It was extracted using [mergekit](https://github.com/arcee-ai/mergekit).
|
| 14 |
+
|
| 15 |
+
## LoRA Details
|
| 16 |
+
|
| 17 |
+
This LoRA adapter was extracted from [Qwen/Qwen3-Coder-30B-A3B-Instruct](https://huggingface.co/Qwen/Qwen3-Coder-30B-A3B-Instruct) and uses [Qwen/Qwen3-30B-A3B-Instruct-2507](https://huggingface.co/Qwen/Qwen3-30B-A3B-Instruct-2507) as a base.
|
| 18 |
+
|
| 19 |
+
### Parameters
|
| 20 |
+
|
| 21 |
+
The following command was used to extract this LoRA adapter:
|
| 22 |
+
|
| 23 |
+
```sh
|
| 24 |
+
/home/dani/qwen3-merges/.venv/bin/mergekit-extract-lora --model Qwen/Qwen3-Coder-30B-A3B-Instruct --base-model Qwen/Qwen3-30B-A3B-Instruct-2507 --out-path output_path
|
| 25 |
+
```
|
adapter_config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"base_model_name_or_path": "Qwen/Qwen3-30B-A3B-Instruct-2507",
|
| 3 |
+
"peft_type": "LORA",
|
| 4 |
+
"use_rslora": false,
|
| 5 |
+
"target_modules": [
|
| 6 |
+
"k_proj",
|
| 7 |
+
"down_proj",
|
| 8 |
+
"up_proj",
|
| 9 |
+
"o_proj",
|
| 10 |
+
"q_proj",
|
| 11 |
+
"v_proj",
|
| 12 |
+
"gate",
|
| 13 |
+
"gate_proj"
|
| 14 |
+
],
|
| 15 |
+
"modules_to_save": [
|
| 16 |
+
"embed_tokens",
|
| 17 |
+
"q_norm",
|
| 18 |
+
"k_norm",
|
| 19 |
+
"input_layernorm",
|
| 20 |
+
"post_attention_layernorm",
|
| 21 |
+
"norm",
|
| 22 |
+
"lm_head"
|
| 23 |
+
],
|
| 24 |
+
"task_type": "CAUSAL_LM",
|
| 25 |
+
"r": 128,
|
| 26 |
+
"lora_alpha": 128,
|
| 27 |
+
"rank_pattern": {},
|
| 28 |
+
"alpha_pattern": {},
|
| 29 |
+
"lora_dropout": 0.0,
|
| 30 |
+
"fan_in_fan_out": false,
|
| 31 |
+
"inference_mode": true
|
| 32 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c2537b1186abe8766cb1bbf334ce997a11005a8026ffdb23aeae1ae355b913d2
|
| 3 |
+
size 14778850784
|