Upload merged Qwen3-4B-Instruct-2507 model (auto-generated README)

README.md

@@ -1,7 +1,7 @@
 ---
 base_model: Qwen/Qwen3-4B-Instruct-2507
 datasets:
-- alfworld_full_success_v2.jsonl
+- alfworld_cleaned_no_upsampling.jsonl
 language:
 - en
 license: apache-2.0
@@ -38,8 +38,8 @@ tool use, and recovery from errors.
 - Method: LoRA (full precision base)
 - Max sequence length: 4096
 - Epochs: 2
-- Learning rate: 1e-06
-- LoRA: r=32, alpha=64
+- Learning rate: 2e-06
+- LoRA: r=64, alpha=128
 
 ## Usage
 
@@ -62,7 +62,7 @@ model = PeftModel.from_pretrained(model, adapter)
 
 ## Sources & Terms (IMPORTANT)
 
-Training data: alfworld_full_success_v2.jsonl
+Training data: alfworld_cleaned_no_upsampling.jsonl
 
 Dataset License: MIT License. This dataset is used and distributed under the terms of the MIT License.
 Compliance: Users must comply with the MIT license (including copyright notice) and the base model's original terms of use.

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:65c584ede773adab21326b84b8200ce44e0afb3718f9fcd2d2d4d69cdab96eab
+oid sha256:56026c0b7a37d4e3801ce44e92a70a77f32fe3b3e7b23e34ce046936364be8f2
 size 4967215360

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:debc4b7fb241a79a01bafa6d5536ab1aa434de74856e17f757947614d5cb9914
+oid sha256:36cec146becb3fdfebdda15469cad6d4717766220d4af6cbc33ae2394f5f4c6b
 size 3077766632