Upload Indo Math Teacher with Gasing method

Modelfile

@@ -0,0 +1,16 @@
+FROM Qwen/Qwen2.5-Math-1.5B
+SYSTEM "You are a math teacher using the Gasing method"
+
+PARAMETER temperature 0.1
+PARAMETER top_p 0.9
+PARAMETER num_predict 200
+
+# Set stop sequences
+TEMPLATE "{ .System }\n\n{ .Prompt }"
+STOP "[/INST]"
+STOP ">>> "
+STOP "
+
+Human:"
+STOP "
+Human:"

README.md

@@ -0,0 +1,71 @@
+---
+language:
+  - id
+  - en
+tags:
+  - gasing-method
+  - math-tutor
+  - indonesian
+  - education
+  - ollama
+license: apache-2.0
+base_model: Qwen/Qwen2.5-Math-1.5B
+---
+
+# Indo Math Teacher - Gasing Method
+
+This model is a fine-tuned version of Qwen/Qwen2.5-Math-1.5B specialized in teaching mathematics using the Gasing method in Bahasa Indonesia.
+
+## Model Description
+
+The Gasing method (Gampang, Asyik, dan Menyenangkan - Easy, Fun, and Enjoyable) is a teaching approach that simplifies complex mathematical concepts through concrete, easy-to-understand steps.
+
+This model can:
+- Solve basic arithmetic problems directly and simply
+- Explain problem-solving steps in a clear manner
+- Use the Gasing method's approach to mathematical education
+
+## Usage with Ollama
+
+To use this model with Ollama:
+
+1. Install [Ollama](https://ollama.com) on your device
+2. Create a Modelfile:
+
+```
+FROM Qwen/Qwen2.5-Math-1.5B
+SYSTEM "You are a math teacher using the Gasing method"
+
+PARAMETER temperature 0.1
+PARAMETER top_p 0.9
+PARAMETER num_predict 200
+
+# Set stop sequences
+TEMPLATE "{ .System }\n\n{ .Prompt }"
+STOP "[/INST]"
+STOP ">>> "
+STOP "\n\nHuman:"
+STOP "\nHuman:"
+```
+
+3. Create the model in Ollama:
+```bash
+ollama create indo_math_teacher -f Modelfile
+```
+
+4. Run the model:
+```bash
+ollama run indo_math_teacher
+```
+
+## Example Usage
+
+**User**: "Jelaskan cara menghitung 8 ditambah 5 menggunakan metode Gasing!"
+
+**Model**: "Simpan 8 di kepala dan 5 di tangan. Pasangan 8 untuk mencapai 10 adalah 2. Ambil 2 dari 5 di tangan. Sekarang di kepala kita punya 10, dan di tangan tersisa 3. Jadi, 10 ditambah 3 sama dengan 13."
+
+## Model Limitations
+
+- This model is specifically trained for mathematical concepts using the Gasing method
+- It works best with arithmetic problems and basic math concepts
+- Responses are optimized for Indonesian language

adapter/0000100_adapters.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e056dbe73eec1bb3c19be89b30d92d4eb87b268f9c405090972d718f185b194
+size 2497239

adapter/0000200_adapters.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:248e50c42815bae15cec2942c60692f7dd1fc0562257c448dc6a3e1ad0397490
+size 2497239

adapter/adapter_config.json

@@ -0,0 +1,30 @@
+{
+    "adapter_path": "./adapters/Qwen/Qwen2.5-Math-1.5B_indo_tutor_20250327_153426",
+    "batch_size": 4,
+    "config": null,
+    "data": "./data/indo_tutor",
+    "fine_tune_type": "lora",
+    "grad_checkpoint": false,
+    "iters": 200,
+    "learning_rate": 5e-06,
+    "lora_parameters": {
+        "rank": 8,
+        "alpha": 16,
+        "dropout": 0.0,
+        "scale": 10.0
+    },
+    "lr_schedule": null,
+    "mask_prompt": false,
+    "max_seq_length": 2048,
+    "model": "Qwen/Qwen2.5-Math-1.5B",
+    "num_layers": 16,
+    "resume_adapter_file": null,
+    "save_every": 100,
+    "seed": 0,
+    "steps_per_eval": 200,
+    "steps_per_report": 10,
+    "test": false,
+    "test_batches": 500,
+    "train": true,
+    "val_batches": 25
+}

adapter/adapters.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:248e50c42815bae15cec2942c60692f7dd1fc0562257c448dc6a3e1ad0397490
+size 2497239

config.json

@@ -0,0 +1,8 @@
+{
+  "base_model": "Qwen/Qwen2.5-Math-1.5B",
+  "model_type": "lora",
+  "system_prompt": "You are a math teacher using the Gasing method",
+  "temperature": 0.1,
+  "top_p": 0.9,
+  "num_predict": 200
+}