Instructions to use bart1259/MiniCOTMath with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use bart1259/MiniCOTMath with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="bart1259/MiniCOTMath")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("bart1259/MiniCOTMath") model = AutoModelForCausalLM.from_pretrained("bart1259/MiniCOTMath") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use bart1259/MiniCOTMath with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "bart1259/MiniCOTMath" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "bart1259/MiniCOTMath", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/bart1259/MiniCOTMath
- SGLang
How to use bart1259/MiniCOTMath 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 "bart1259/MiniCOTMath" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "bart1259/MiniCOTMath", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'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 "bart1259/MiniCOTMath" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "bart1259/MiniCOTMath", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use bart1259/MiniCOTMath with Docker Model Runner:
docker model run hf.co/bart1259/MiniCOTMath
Upload folder using huggingface_hub
Browse files- README.md +12 -11
- model.safetensors +1 -1
- score.png +0 -0
README.md
CHANGED
|
@@ -10,8 +10,11 @@ Model details:
|
|
| 10 |
- **Attention Head Count**: 4
|
| 11 |
- **Residual Stream Size**: 256
|
| 12 |
- **Context Length**: 256
|
| 13 |
-
- **Tokens Trained on**: 419,
|
| 14 |
|
|
|
|
|
|
|
|
|
|
| 15 |
|
| 16 |
```py
|
| 17 |
from transformers import pipeline
|
|
@@ -19,7 +22,8 @@ from transformers import pipeline
|
|
| 19 |
pipe = pipeline(
|
| 20 |
"text-generation", model="bart1259/MiniCOTMath"
|
| 21 |
)
|
| 22 |
-
print(pipe("Input: (5 + 5)
|
|
|
|
| 23 |
```
|
| 24 |
|
| 25 |
Outputs:
|
|
@@ -59,7 +63,8 @@ class StopCriteria(StoppingCriteria):
|
|
| 59 |
def __iter__(self):
|
| 60 |
yield self
|
| 61 |
|
| 62 |
-
prompt = "Input: (
|
|
|
|
| 63 |
|
| 64 |
tokenizer = AutoTokenizer.from_pretrained("bart1259/MiniCOTMath")
|
| 65 |
model = AutoModelForCausalLM.from_pretrained("bart1259/MiniCOTMath").cuda()
|
|
@@ -81,18 +86,14 @@ _ = model.generate(
|
|
| 81 |
|
| 82 |
Outputs:
|
| 83 |
```
|
| 84 |
-
Input: (
|
| 85 |
|
| 86 |
Step 1:
|
| 87 |
-
(
|
| 88 |
(5 + 5) = 10
|
| 89 |
|
| 90 |
Step 2:
|
| 91 |
-
|
| 92 |
-
|
| 93 |
-
|
| 94 |
-
Step 3:
|
| 95 |
-
30
|
| 96 |
-
Final Result: 30
|
| 97 |
<end>
|
| 98 |
```
|
|
|
|
| 10 |
- **Attention Head Count**: 4
|
| 11 |
- **Residual Stream Size**: 256
|
| 12 |
- **Context Length**: 256
|
| 13 |
+
- **Tokens Trained on**: 419,649,024
|
| 14 |
|
| 15 |
+
Training Score During Training
|
| 16 |
+
|
| 17 |
+
[score.png](score.png)
|
| 18 |
|
| 19 |
```py
|
| 20 |
from transformers import pipeline
|
|
|
|
| 22 |
pipe = pipeline(
|
| 23 |
"text-generation", model="bart1259/MiniCOTMath"
|
| 24 |
)
|
| 25 |
+
print(pipe("Input: (5 + 5)
|
| 26 |
+
", max_new_tokens=100)[0]["generated_text"])
|
| 27 |
```
|
| 28 |
|
| 29 |
Outputs:
|
|
|
|
| 63 |
def __iter__(self):
|
| 64 |
yield self
|
| 65 |
|
| 66 |
+
prompt = "Input: (5 + 5)
|
| 67 |
+
"
|
| 68 |
|
| 69 |
tokenizer = AutoTokenizer.from_pretrained("bart1259/MiniCOTMath")
|
| 70 |
model = AutoModelForCausalLM.from_pretrained("bart1259/MiniCOTMath").cuda()
|
|
|
|
| 86 |
|
| 87 |
Outputs:
|
| 88 |
```
|
| 89 |
+
Input: (5 + 5)
|
| 90 |
|
| 91 |
Step 1:
|
| 92 |
+
(5 + 5)
|
| 93 |
(5 + 5) = 10
|
| 94 |
|
| 95 |
Step 2:
|
| 96 |
+
10
|
| 97 |
+
Final Result: 10
|
|
|
|
|
|
|
|
|
|
|
|
|
| 98 |
<end>
|
| 99 |
```
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 25352072
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:80470ec0e1be3e0e2ae474707fdd150d02d666915dfaa5d54528418c03e7da66
|
| 3 |
size 25352072
|
score.png
ADDED
|