Instructions to use cs-552-2026-OAAA/math_model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use cs-552-2026-OAAA/math_model with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="cs-552-2026-OAAA/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-OAAA/math_model") model = AutoModelForCausalLM.from_pretrained("cs-552-2026-OAAA/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-OAAA/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-OAAA/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-OAAA/math_model", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/cs-552-2026-OAAA/math_model
- SGLang
How to use cs-552-2026-OAAA/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-OAAA/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-OAAA/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-OAAA/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-OAAA/math_model", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use cs-552-2026-OAAA/math_model with Docker Model Runner:
docker model run hf.co/cs-552-2026-OAAA/math_model
Automated MNLP evaluation report (2026-06-11)
#1
by zechen-nlp - opened
- EVAL_REPORT.md +178 -0
EVAL_REPORT.md
ADDED
|
@@ -0,0 +1,178 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Automated MNLP evaluation report
|
| 2 |
+
|
| 3 |
+
- **Model repo:** [`cs-552-2026-OAAA/math_model`](https://huggingface.co/cs-552-2026-OAAA/math_model)
|
| 4 |
+
- **Owner(s):** group **OAAA**
|
| 5 |
+
- **Generated at:** 2026-06-11T06:23:10+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._
|
| 9 |
+
|
| 10 |
+
## Evaluated checkpoint
|
| 11 |
+
|
| 12 |
+
- **Commit:** [`a7fc550`](https://huggingface.co/cs-552-2026-OAAA/math_model/commit/a7fc55016982451246691a3530624ef17acea9a2)
|
| 13 |
+
- **Message:** Upload generation_config.json with huggingface_hub
|
| 14 |
+
- **Committed:** 2026-06-07T14:19:42+00:00
|
| 15 |
+
|
| 16 |
+
## Summary
|
| 17 |
+
|
| 18 |
+
| Benchmark | Accuracy | Status |
|
| 19 |
+
|---|---:|---|
|
| 20 |
+
| Math | 0.4800 | ok |
|
| 21 |
+
| Knowledge | — | not run |
|
| 22 |
+
| Multilingual | — | not run |
|
| 23 |
+
| Safety | — | not run |
|
| 24 |
+
|
| 25 |
+
## Sample completions
|
| 26 |
+
|
| 27 |
+
_Prompts are intentionally omitted to avoid revealing benchmark contents. For multi-completion problems, only one completion is shown per sample._
|
| 28 |
+
|
| 29 |
+
### Math
|
| 30 |
+
|
| 31 |
+
**Correct** (1 shown)
|
| 32 |
+
|
| 33 |
+
- **reference**: `1012`
|
| 34 |
+
- **overall** (6/8 completions correct)
|
| 35 |
+
- **extracted** (✓): `1012`
|
| 36 |
+
- **completion**:
|
| 37 |
+
|
| 38 |
+
```text
|
| 39 |
+
<think>
|
| 40 |
+
Okay, so I need to find the maximum possible value of this sum:
|
| 41 |
+
|
| 42 |
+
$$
|
| 43 |
+
\sum_{i=1}^{2023} \frac{a_i - a_{i-1}}{a_{i+1} - a_i}
|
| 44 |
+
$$
|
| 45 |
+
|
| 46 |
+
given that $a_1, \ldots, a_{2024}$ are positive real numbers with $a_{i+1} \ge 2a_i$ for each $i = 1, \ldots, 2023$, and $a_0 = 0$.
|
| 47 |
+
|
| 48 |
+
First, let me try to understand the problem. We have a sequence starting at 0, and each term is at least twice the previous one. So,
|
| 49 |
+
|
| 50 |
+
... [truncated 822 chars] ...
|
| 51 |
+
|
| 52 |
+
4a, \quad \ldots, \quad a_{2024} = 2^{2023}a
|
| 53 |
+
$$
|
| 54 |
+
|
| 55 |
+
This is a geometric progression with ratio 2.
|
| 56 |
+
|
| 57 |
+
---
|
| 58 |
+
|
| 59 |
+
### Step 3: Computing the Sum in the Optimal Case
|
| 60 |
+
|
| 61 |
+
Let’s compute the sum when $ a_{i+1} = 2a_i $ for all $ i $, which gives:
|
| 62 |
+
|
| 63 |
+
$$
|
| 64 |
+
\frac{a_i - a_{i-1}}{a_{i+1} - a_i} = \frac{a_i - a_{i-1}}{a_i} = 1 - \frac{a_{i-1}}{a_i}
|
| 65 |
+
$$
|
| 66 |
+
|
| 67 |
+
So the entire sum becomes:
|
| 68 |
+
|
| 69 |
+
$$
|
| 70 |
+
\sum_{i=1}^{2023} \left(1 - \frac{a_{i-1}}{a_i} \right) = 2023 - \sum_{i=1}^{2023} \frac{a_{i-1}}{a_i}
|
| 71 |
+
$$
|
| 72 |
+
|
| 73 |
+
Now, compute the second term:
|
| 74 |
+
|
| 75 |
+
$$
|
| 76 |
+
\sum_{i=1}^{2023} \frac{a_{i-1}}{a_i} = \frac{a_0}{a_1} + \frac{a_1}{a_2} + \frac{a_2}{a_3} + \cdots + \frac{a_{2023}}{a_{2024}}
|
| 77 |
+
$$
|
| 78 |
+
|
| 79 |
+
Since $ a_{i} = 2^{i-1}a_1 $, we have:
|
| 80 |
+
|
| 81 |
+
$$
|
| 82 |
+
\frac{a_{i-1}}{a_i} = \frac{2^{i-2}a_1}{2^{i-1}a_1} = \frac{1}{2}
|
| 83 |
+
$$
|
| 84 |
+
|
| 85 |
+
This holds for all $ i \ge 2 $, and the first term is $ \frac{a_0}{a_1} = 0 $.
|
| 86 |
+
|
| 87 |
+
So the sum becomes:
|
| 88 |
+
|
| 89 |
+
$$
|
| 90 |
+
\sum_{i=1}^{2023} \frac{a_{i-1}}{a_i} = 0 + 2022 \cdot \frac{1}{2} = 1011
|
| 91 |
+
$$
|
| 92 |
+
|
| 93 |
+
Thus, the total sum is:
|
| 94 |
+
|
| 95 |
+
$$
|
| 96 |
+
2023 - 1011 = \boxed{1012}
|
| 97 |
+
$$
|
| 98 |
+
|
| 99 |
+
---
|
| 100 |
+
|
| 101 |
+
### Final Conclusion
|
| 102 |
+
|
| 103 |
+
The maximum possible value of the given sum is achieved when the sequence grows exponentially with a ratio of 2 at each step. In this optimal case, the sum evaluates to:
|
| 104 |
+
|
| 105 |
+
$$
|
| 106 |
+
\boxed{1012}
|
| 107 |
+
$$
|
| 108 |
+
```
|
| 109 |
+
|
| 110 |
+
**Incorrect** (1 shown)
|
| 111 |
+
|
| 112 |
+
- **reference**: `237`
|
| 113 |
+
- **overall** (0/8 completions correct)
|
| 114 |
+
- **extracted** (✗): `60671`
|
| 115 |
+
- **completion**:
|
| 116 |
+
|
| 117 |
+
```text
|
| 118 |
+
<think>
|
| 119 |
+
Okay, so I need to solve this problem where I have to find the probability that a randomly chosen non-empty subset B of the positive divisors of 2025 has a least common multiple (LCM) equal to 2025. Then express that probability as a reduced fraction m/n and find m + n. Alright, let's start by understanding the problem step by step.
|
| 120 |
+
|
| 121 |
+
First, let me recall that the divisors of a number can b
|
| 122 |
+
|
| 123 |
+
... [truncated 822 chars] ...
|
| 124 |
+
|
| 125 |
+
least one** element from the set of elements with $ a = 4 $ **and** at least one from the set with $ b = 2 $.
|
| 126 |
+
|
| 127 |
+
Let’s denote:
|
| 128 |
+
|
| 129 |
+
- $ N $: the number of subsets of $ A $ that contain **at least one** element with $ a = 4 $ and **at least one** element with $ b = 2 $
|
| 130 |
+
- The number of such subsets is given by the inclusion-exclusion principle:
|
| 131 |
+
|
| 132 |
+
$$
|
| 133 |
+
N = 2^{15} - 2^{12} - 2^{10} + 2^8
|
| 134 |
+
$$
|
| 135 |
+
|
| 136 |
+
$$
|
| 137 |
+
N = 32768 - 4096 - 1024 + 256 = 27904
|
| 138 |
+
$$
|
| 139 |
+
|
| 140 |
+
This counts **all subsets** (including the empty set) that contain at least one element with $ a = 4 $ and at least one with $ b = 2 $. Since the **empty set** is not in this set, the number of **non-empty** subsets is:
|
| 141 |
+
|
| 142 |
+
$$
|
| 143 |
+
N = 27904
|
| 144 |
+
$$
|
| 145 |
+
|
| 146 |
+
---
|
| 147 |
+
|
| 148 |
+
### Step 4: Compute the Probability
|
| 149 |
+
|
| 150 |
+
The probability is:
|
| 151 |
+
|
| 152 |
+
$$
|
| 153 |
+
\frac{27904}{32767}
|
| 154 |
+
$$
|
| 155 |
+
|
| 156 |
+
We now check whether this fraction is in its lowest terms.
|
| 157 |
+
|
| 158 |
+
- $ 32767 = 7 \cdot 31 \cdot 151 $
|
| 159 |
+
- $ 27904 = 2^8 \cdot 109 $
|
| 160 |
+
|
| 161 |
+
Since 109 is a prime number and does **not** divide 32767, and 2 is not a factor of 32767, the two numbers are **coprime**.
|
| 162 |
+
|
| 163 |
+
Thus, the fraction is in its simplest form:
|
| 164 |
+
|
| 165 |
+
$$
|
| 166 |
+
\frac{27904}{32767}
|
| 167 |
+
$$
|
| 168 |
+
|
| 169 |
+
---
|
| 170 |
+
|
| 171 |
+
### Final Step: Compute $ m + n $
|
| 172 |
+
|
| 173 |
+
Here, $ m = 27904 $, $ n = 32767 $, so:
|
| 174 |
+
|
| 175 |
+
$$
|
| 176 |
+
m + n = 27904 + 32767 = \boxed{60671}
|
| 177 |
+
$$
|
| 178 |
+
```
|