Text Generation
Transformers
Safetensors
qwen3
Generated from Trainer
sft
trl
conversational
text-generation-inference
Instructions to use cs-552-2026-flab/math_model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use cs-552-2026-flab/math_model with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="cs-552-2026-flab/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-flab/math_model") model = AutoModelForCausalLM.from_pretrained("cs-552-2026-flab/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-flab/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-flab/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-flab/math_model", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/cs-552-2026-flab/math_model
- SGLang
How to use cs-552-2026-flab/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-flab/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-flab/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-flab/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-flab/math_model", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use cs-552-2026-flab/math_model with Docker Model Runner:
docker model run hf.co/cs-552-2026-flab/math_model
Update Automated MNLP evaluation report (2026-05-23)
Browse files- EVAL_REPORT.md +44 -34
EVAL_REPORT.md
CHANGED
|
@@ -2,7 +2,7 @@
|
|
| 2 |
|
| 3 |
- **Model repo:** [`cs-552-2026-flab/math_model`](https://huggingface.co/cs-552-2026-flab/math_model)
|
| 4 |
- **Owner(s):** group **flab**
|
| 5 |
-
- **Generated at:** 2026-05-
|
| 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 |
|
|
@@ -24,62 +24,72 @@ _Prompts are intentionally omitted to avoid revealing benchmark contents. For mu
|
|
| 24 |
|
| 25 |
**Correct** (1 shown)
|
| 26 |
|
| 27 |
-
- **reference**: `
|
| 28 |
-
- **overall** (
|
| 29 |
-
- **extracted** (✓): `
|
| 30 |
- **completion**:
|
| 31 |
|
| 32 |
```text
|
| 33 |
</think>
|
| 34 |
|
| 35 |
-
The
|
| 36 |
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
|
| 40 |
-
If \(k \le 193\), then the board can be tiled with \(97\) dominoes (each covering two cells). Since each domino can contain at most one marked cell, at most \(97\) of these dominoes can be chosen to contain a marked cell. The remaining \(194 - 97 = 97\) dominoes can be tiled without any
|
| 41 |
-
|
| 42 |
-
... [truncated 646 chars] ...
|
| 43 |
-
|
| 44 |
-
inoes do not contain a marked cell.
|
| 45 |
-
|
| 46 |
-
- **Case \(n = 194\):**
|
| 47 |
-
Split the board into \(97\) disjoint \(2 \times 2\) squares. In each square, choose one cell to mark and mark the other cell of the same domino. This uses \(97\) marked cells.
|
| 48 |
-
Tile the board with dominoes:
|
| 49 |
-
- In each \(2 \times 2\) square, tile it with two vertical dominoes (or two horizontal dominoes).
|
| 50 |
-
- Combine the vertical dominoes from adjacent squares to form a larger horizontal domino, and combine the horizontal dominoes to form a larger vertical domino.
|
| 51 |
-
The result is a single \(194 \times 194\) board tiled by \(97\) dominoes (each covering two cells). The chosen marked cell belongs to exactly one domino, and all other dominoes contain no marked cells.
|
| 52 |
-
This tiling uses exactly one marked cell, so \(k = 1\). Hence \(k = 97\) is attainable.
|
| 53 |
|
| 54 |
-
|
| 55 |
|
| 56 |
-
|
| 57 |
-
The minimum possible value of \(k\) is \(97\).
|
| 58 |
```
|
| 59 |
|
| 60 |
**Incorrect** (1 shown)
|
| 61 |
|
| 62 |
-
- **reference**: `
|
| 63 |
- **overall** (0/8 completions correct)
|
| 64 |
-
- **extracted** (✗): `
|
| 65 |
- **completion**:
|
| 66 |
|
| 67 |
```text
|
| 68 |
</think>
|
| 69 |
|
| 70 |
-
|
| 71 |
-
|
| 72 |
\[
|
| 73 |
-
|
|
|
|
| 74 |
\]
|
| 75 |
|
| 76 |
-
|
|
|
|
|
|
|
|
|
|
| 77 |
|
|
|
|
|
|
|
|
|
|
| 78 |
\[
|
| 79 |
-
a
|
| 80 |
-
\]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 81 |
|
| 82 |
\[
|
| 83 |
-
\boxed{
|
| 84 |
\]
|
| 85 |
```
|
|
|
|
| 2 |
|
| 3 |
- **Model repo:** [`cs-552-2026-flab/math_model`](https://huggingface.co/cs-552-2026-flab/math_model)
|
| 4 |
- **Owner(s):** group **flab**
|
| 5 |
+
- **Generated at:** 2026-05-23T14:37:35+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 | 0.3400 | ok |
|
| 15 |
| Knowledge | — | not run |
|
| 16 |
| Multilingual | — | not run |
|
| 17 |
| Safety | — | not run |
|
|
|
|
| 24 |
|
| 25 |
**Correct** (1 shown)
|
| 26 |
|
| 27 |
+
- **reference**: `\frac{448}{15625}`
|
| 28 |
+
- **overall** (4/8 completions correct)
|
| 29 |
+
- **extracted** (✓): `\frac{448}{15625}`
|
| 30 |
- **completion**:
|
| 31 |
|
| 32 |
```text
|
| 33 |
</think>
|
| 34 |
|
| 35 |
+
The probability that exactly 4 out of the 7 islands contain treasure is given by the binomial distribution:
|
| 36 |
|
| 37 |
+
\[
|
| 38 |
+
P(X = 4) = \binom{7}{4} \left(\frac{1}{5}\right)^4 \left(\frac{4}{5}\right)^3 = 35 \cdot \frac{1}{625} \cdot \frac{64}{125} = \frac{2240}{78125}.
|
| 39 |
+
\]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
|
| 41 |
+
Simplifying the fraction by dividing numerator and denominator by 5 gives \(\frac{448}{15625}\).
|
| 42 |
|
| 43 |
+
Thus, the required probability is \(\boxed{\frac{448}{15625}}\).
|
|
|
|
| 44 |
```
|
| 45 |
|
| 46 |
**Incorrect** (1 shown)
|
| 47 |
|
| 48 |
+
- **reference**: `$g(x)=c, g(x)=\lceil x \rceil, g(x)=\lfloor x \rfloor$`
|
| 49 |
- **overall** (0/8 completions correct)
|
| 50 |
+
- **extracted** (✗): `h(x)=c\ \text{for some integer }c`
|
| 51 |
- **completion**:
|
| 52 |
|
| 53 |
```text
|
| 54 |
</think>
|
| 55 |
|
| 56 |
+
We need to determine all functions \(h:\mathbb{Q}\to\mathbb{Z}\) satisfying
|
|
|
|
| 57 |
\[
|
| 58 |
+
h(x)=h\!\left(\frac{h(bx-a)+a}{b}\right)
|
| 59 |
+
\qquad\text{for all }x\in\mathbb{Q},\;a\in\mathbb{Z},\;b\in\mathbb{Z}^+\ .
|
| 60 |
\]
|
| 61 |
|
| 62 |
+
**Step 1. Constant solutions.**
|
| 63 |
+
If \(h(x)=c\) for all \(x\in\mathbb{Q}\) and any integer \(c\), then the equation holds because both sides are \(c\). Hence all constant functions are sol
|
| 64 |
+
|
| 65 |
+
... [truncated 110 chars] ...
|
| 66 |
|
| 67 |
+
\(h(x)\neq h(bx-a)+a\) (otherwise the constant solution would also work).
|
| 68 |
+
|
| 69 |
+
Take \(x=0\). For any \(a,b\in\mathbb{Z}^+\),
|
| 70 |
\[
|
| 71 |
+
h(0)=h\!\left(\frac{h(b\cdot0-a)+a}{b}\right)=h\!\left(\frac{-h(a)+a}{b}\right).
|
| 72 |
+
\]
|
| 73 |
+
If \(h(0)=c\), then \(\frac{-h(a)+a}{b}=c\) for all \(a,b\in\mathbb{Z}^+\), i.e., \(-h(a)+a = bc\).
|
| 74 |
+
If \(b=1\) this gives \(a-h(a)=c\), so \(h(a)=a-c\); then \(h(0)=0-c=-c\).
|
| 75 |
+
Thus \(h(a)=a-c\) for all \(a\in\mathbb{Z}\).
|
| 76 |
+
|
| 77 |
+
Now use \(h(a)=a-c\) in the original equation. Let \(b=1\). For any \(x\in\mathbb{Q}\),
|
| 78 |
+
\[
|
| 79 |
+
h(x)=h\!\left(\frac{h(x-a)+a}{1}\right)=h(h(x-a)+a).
|
| 80 |
+
\]
|
| 81 |
+
Since \(h(h(x-a)+a)=h(x-a)+a\), we obtain \(h(x)=h(x-a)+a\) for all \(x\in\mathbb{Q}\), \(a\in\mathbb{Z}\).
|
| 82 |
+
|
| 83 |
+
Take any \(a\in\mathbb{Z}\). For \(x=0\) this gives \(h(0)=h(-a)+a\).
|
| 84 |
+
For \(x=a\) we have \(h(a)=h(a-a)+a=h(0)+a\), so \(h(a)=h(0)+a\).
|
| 85 |
+
Thus \(h(0)=h(a)\) for all \(a\in\mathbb{Z}\). Hence \(h\) is constant.
|
| 86 |
+
|
| 87 |
+
Therefore the only non‑constant solutions are the constant functions.
|
| 88 |
+
|
| 89 |
+
**Conclusion.**
|
| 90 |
+
All functions \(h:\mathbb{Q}\to\mathbb{Z}\) satisfying the given equation are exactly the constant functions \(h(x)=c\) for some integer \(c\).
|
| 91 |
|
| 92 |
\[
|
| 93 |
+
\boxed{h(x)=c\ \text{for some integer }c}
|
| 94 |
\]
|
| 95 |
```
|