Text Generation
PEFT
Safetensors
Transformers
qwen3
grpo
lora
trl
conversational
text-generation-inference
Instructions to use cs-552-2026-thinking-tokens/math_model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use cs-552-2026-thinking-tokens/math_model with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-1.7B") model = PeftModel.from_pretrained(base_model, "cs-552-2026-thinking-tokens/math_model") - Transformers
How to use cs-552-2026-thinking-tokens/math_model with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="cs-552-2026-thinking-tokens/math_model") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("cs-552-2026-thinking-tokens/math_model") model = AutoModelForCausalLM.from_pretrained("cs-552-2026-thinking-tokens/math_model", device_map="auto") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=40) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use cs-552-2026-thinking-tokens/math_model with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "cs-552-2026-thinking-tokens/math_model" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "cs-552-2026-thinking-tokens/math_model", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/cs-552-2026-thinking-tokens/math_model
- SGLang
How to use cs-552-2026-thinking-tokens/math_model with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "cs-552-2026-thinking-tokens/math_model" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "cs-552-2026-thinking-tokens/math_model", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "cs-552-2026-thinking-tokens/math_model" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "cs-552-2026-thinking-tokens/math_model", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use cs-552-2026-thinking-tokens/math_model with Docker Model Runner:
docker model run hf.co/cs-552-2026-thinking-tokens/math_model
Update Automated MNLP evaluation report (2026-06-05)
Browse files- EVAL_REPORT.md +32 -31
EVAL_REPORT.md
CHANGED
|
@@ -2,7 +2,7 @@
|
|
| 2 |
|
| 3 |
- **Model repo:** [`cs-552-2026-thinking-tokens/math_model`](https://huggingface.co/cs-552-2026-thinking-tokens/math_model)
|
| 4 |
- **Owner(s):** group **thinking-tokens**
|
| 5 |
-
- **Generated at:** 2026-06-
|
| 6 |
- **Pipeline:** [mnlp-project-ci](https://github.com/eric11eca/mnlp-project-ci)
|
| 7 |
|
| 8 |
_This PR is opened automatically by the course CI. It is **non-blocking** — you do not need to merge it. The next nightly run will refresh this file._
|
|
@@ -11,7 +11,7 @@ _This PR is opened automatically by the course CI. It is **non-blocking** — yo
|
|
| 11 |
|
| 12 |
| Benchmark | Accuracy | Status |
|
| 13 |
|---|---:|---|
|
| 14 |
-
| Math | 0.
|
| 15 |
| Knowledge | — | not run |
|
| 16 |
| Multilingual | — | not run |
|
| 17 |
| Safety | — | not run |
|
|
@@ -40,7 +40,9 @@ _Prompts are intentionally omitted to avoid revealing benchmark contents. For mu
|
|
| 40 |
|
| 41 |
... [truncated 823 chars] ...
|
| 42 |
|
| 43 |
-
|
|
|
|
|
|
|
| 44 |
|
| 45 |
---
|
| 46 |
|
|
@@ -58,15 +60,17 @@ _Prompts are intentionally omitted to avoid revealing benchmark contents. For mu
|
|
| 58 |
|
| 59 |
### Step 6: Final Answer
|
| 60 |
|
| 61 |
-
This is a known result in combinatorics. For a $n \times n$ board, the **minimum number of marked cells** required to ensure that **only one tiling**
|
| 62 |
|
| 63 |
$$
|
| 64 |
\boxed{194}
|
| 65 |
$$
|
| 66 |
|
| 67 |
-
This is
|
|
|
|
|
|
|
| 68 |
|
| 69 |
-
|
| 70 |
|
| 71 |
$$
|
| 72 |
\boxed{194}
|
|
@@ -77,7 +81,7 @@ _Prompts are intentionally omitted to avoid revealing benchmark contents. For mu
|
|
| 77 |
|
| 78 |
- **reference**: `$g(x)=c, g(x)=\lceil x \rceil, g(x)=\lfloor x \rfloor$`
|
| 79 |
- **overall** (0/8 completions correct)
|
| 80 |
-
- **extracted** (✗): `h(x) = c \text{ for some } c \in \mathbb{Z}
|
| 81 |
- **completion**:
|
| 82 |
|
| 83 |
```text
|
|
@@ -103,18 +107,7 @@ _Prompts are intentionally omitted to avoid revealing benchmark contents. For mu
|
|
| 103 |
|
| 104 |
... [truncated 822 chars] ...
|
| 105 |
|
| 106 |
-
|
| 107 |
-
$$
|
| 108 |
-
|
| 109 |
-
Matching coefficients:
|
| 110 |
-
|
| 111 |
-
- Coefficient of $ x $: $ k = k^2 \Rightarrow k = 0 $ or $ k = 1 $
|
| 112 |
-
- Constant term: $ m = \frac{-k^2a + km + ka}{b} + m $
|
| 113 |
-
|
| 114 |
-
Try $ k = 1 $:
|
| 115 |
-
|
| 116 |
-
- Coefficient of $ x $: OK
|
| 117 |
-
- Constant term: $ m = \frac{-1^2a + 1\cdot m + 1\cdot a}{b} + m = \frac{-a + m + a}{b} + m = \frac{m}{b} + m $
|
| 118 |
|
| 119 |
So:
|
| 120 |
|
|
@@ -124,16 +117,17 @@ _Prompts are intentionally omitted to avoid revealing benchmark contents. For mu
|
|
| 124 |
|
| 125 |
So $ h(x) = x $ is a solution.
|
| 126 |
|
| 127 |
-
|
| 128 |
|
| 129 |
$$
|
| 130 |
-
h(x) = x
|
| 131 |
$$
|
| 132 |
|
| 133 |
-
|
| 134 |
|
| 135 |
$$
|
| 136 |
-
h(
|
|
|
|
| 137 |
$$
|
| 138 |
|
| 139 |
So $ h(x) = x $ is a solution.
|
|
@@ -142,22 +136,29 @@ _Prompts are intentionally omitted to avoid revealing benchmark contents. For mu
|
|
| 142 |
|
| 143 |
### Step 5: Are there other solutions?
|
| 144 |
|
| 145 |
-
We have found:
|
|
|
|
|
|
|
|
|
|
| 146 |
|
| 147 |
-
|
| 148 |
-
- Linear function: $ h(x) = x $
|
| 149 |
|
| 150 |
-
|
| 151 |
|
| 152 |
-
Suppose $ h
|
| 153 |
|
| 154 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 155 |
|
| 156 |
---
|
| 157 |
|
| 158 |
-
###
|
| 159 |
|
| 160 |
$$
|
| 161 |
-
\boxed{h(x) = c \text{ for some } c \in \mathbb{Z}
|
| 162 |
$$
|
| 163 |
```
|
|
|
|
| 2 |
|
| 3 |
- **Model repo:** [`cs-552-2026-thinking-tokens/math_model`](https://huggingface.co/cs-552-2026-thinking-tokens/math_model)
|
| 4 |
- **Owner(s):** group **thinking-tokens**
|
| 5 |
+
- **Generated at:** 2026-06-05T03:00:32+00:00 (UTC)
|
| 6 |
- **Pipeline:** [mnlp-project-ci](https://github.com/eric11eca/mnlp-project-ci)
|
| 7 |
|
| 8 |
_This PR is opened automatically by the course CI. It is **non-blocking** — you do not need to merge it. The next nightly run will refresh this file._
|
|
|
|
| 11 |
|
| 12 |
| Benchmark | Accuracy | Status |
|
| 13 |
|---|---:|---|
|
| 14 |
+
| Math | 0.4000 | ok |
|
| 15 |
| Knowledge | — | not run |
|
| 16 |
| Multilingual | — | not run |
|
| 17 |
| Safety | — | not run |
|
|
|
|
| 40 |
|
| 41 |
... [truncated 823 chars] ...
|
| 42 |
|
| 43 |
+
t **every domino contains at most one marked cell**.
|
| 44 |
+
|
| 45 |
+
This is a **combinatorial problem** where we want to **minimize the number of marked cells** such that **only one tiling** satisfies the condition.
|
| 46 |
|
| 47 |
---
|
| 48 |
|
|
|
|
| 60 |
|
| 61 |
### Step 6: Final Answer
|
| 62 |
|
| 63 |
+
This is a known result in combinatorics. For a $n \times n$ board, the **minimum number of marked cells** required to ensure that **only one tiling** of the board with $1 \times 2$ dominoes has **at most one marked cell per domino** is:
|
| 64 |
|
| 65 |
$$
|
| 66 |
\boxed{194}
|
| 67 |
$$
|
| 68 |
|
| 69 |
+
This is the **minimum** number of marked cells required to ensure that **only one tiling** satisfies the condition.
|
| 70 |
+
|
| 71 |
+
---
|
| 72 |
|
| 73 |
+
### ✅ Final Answer:
|
| 74 |
|
| 75 |
$$
|
| 76 |
\boxed{194}
|
|
|
|
| 81 |
|
| 82 |
- **reference**: `$g(x)=c, g(x)=\lceil x \rceil, g(x)=\lfloor x \rfloor$`
|
| 83 |
- **overall** (0/8 completions correct)
|
| 84 |
+
- **extracted** (✗): `h(x) = c \text{ or } h(x) = x \text{ for some } c \in \mathbb{Z}`
|
| 85 |
- **completion**:
|
| 86 |
|
| 87 |
```text
|
|
|
|
| 107 |
|
| 108 |
... [truncated 822 chars] ...
|
| 109 |
|
| 110 |
+
- a)}{b} + m = - \frac{-m}{b} + m = \frac{m}{b} + m $
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 111 |
|
| 112 |
So:
|
| 113 |
|
|
|
|
| 117 |
|
| 118 |
So $ h(x) = x $ is a solution.
|
| 119 |
|
| 120 |
+
Let’s verify:
|
| 121 |
|
| 122 |
$$
|
| 123 |
+
h(x) = x \Rightarrow h(bx - a) = bx - a
|
| 124 |
$$
|
| 125 |
|
| 126 |
+
Then:
|
| 127 |
|
| 128 |
$$
|
| 129 |
+
\frac{h(bx - a) + a}{b} = \frac{bx - a + a}{b} = x
|
| 130 |
+
\Rightarrow h(x) = x
|
| 131 |
$$
|
| 132 |
|
| 133 |
So $ h(x) = x $ is a solution.
|
|
|
|
| 136 |
|
| 137 |
### Step 5: Are there other solutions?
|
| 138 |
|
| 139 |
+
We have found two types of solutions:
|
| 140 |
+
|
| 141 |
+
1. **Constant functions**: $ h(x) = c $, for some $ c \in \mathbb{Z} $
|
| 142 |
+
2. **Linear functions**: $ h(x) = x $, which is also a solution
|
| 143 |
|
| 144 |
+
Are there any other solutions?
|
|
|
|
| 145 |
|
| 146 |
+
Suppose $ h $ is not constant or linear. Let’s try $ h(x) = c $, a constant function, and $ h(x) = x $, and see if they are the **only** solutions.
|
| 147 |
|
| 148 |
+
Suppose $ h $ is not constant, and not linear. Then we must show that it must be constant or linear.
|
| 149 |
|
| 150 |
+
But from the earlier analysis, **no non-constant linear functions** work, and **constant functions** and $ h(x) = x $ both work.
|
| 151 |
+
|
| 152 |
+
So the only possible solutions are:
|
| 153 |
+
|
| 154 |
+
- Constant functions: $ h(x) = c $, for some $ c \in \mathbb{Z} $
|
| 155 |
+
- $ h(x) = x $
|
| 156 |
|
| 157 |
---
|
| 158 |
|
| 159 |
+
### Final Answer:
|
| 160 |
|
| 161 |
$$
|
| 162 |
+
\boxed{h(x) = c \text{ or } h(x) = x \text{ for some } c \in \mathbb{Z}}
|
| 163 |
$$
|
| 164 |
```
|