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

README.md

@@ -17,6 +17,7 @@ tags:
 - agentbench
 - alfworld
 - dbbench
+- db-oversampling
 ---
 
 # qwen3-4b-agentbench-dbalf-lora
@@ -45,29 +46,29 @@ Loss is applied to **all assistant turns** in the trajectory, enabling the model
 
 - DBBench dataset: `u-10bei/dbbench_sft_dataset_react_v4`
 - ALFWorld dataset: `u-10bei/sft_alfworld_trajectory_dataset_v5`
-- Mixing ratio (pre-merge target): **DB:ALF = 2:1**
+- Mixing ratio (pre-merge target): **DB:ALF = 1:0**
 
 ### DB Oversampling (category-aware)
-Enabled: **False**
+Enabled: **True**
 
 DB category weights used during training-data preparation:
 
-- counting: 1
-- comparison: 1
+- counting: 2
+- comparison: 2
 - ranking: 1
 - select: 1
-- insert: 1
+- insert: 2
 - update: 1
-- other: 1
+- other: 2
 
 ## Training Configuration
 
 - Base model: Qwen/Qwen3-4B-Instruct-2507
 - Method: LoRA (full precision base)
 - Max sequence length: 2048
-- Epochs: 1
-- Learning rate: 2e-06
-- LoRA: r=32, alpha=64, dropout=0.05
+- Epochs: 1.2
+- Learning rate: 3e-06
+- LoRA: r=32, alpha=64, dropout=0.0
 - Per-device train batch size: 2
 - Gradient accumulation: 4
 

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b2031c7b5bda1d270c62e8ee556b5ddd1a2d990d7899eff7f4ef8c2d10f863d1
+oid sha256:9f7543658eb9e5119afcfd0eae26db7c8892e5f5d56417c552d3377bea4caf8f
 size 4967215360

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e33faeee578db08f9d17cd701de66faf60543891a5cd000f0b21c899f52ded10
+oid sha256:8ae0228ca45d1a58e552ca0c34de7d27287f73be9bd6354f9adf1efa4fc235b3
 size 3077766632