Text Generation
Transformers
Safetensors
Korean
llama
Merge
conversational
Eval Results (legacy)
text-generation-inference
Instructions to use kuotient/EEVE-Instruct-Math-10.8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use kuotient/EEVE-Instruct-Math-10.8B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="kuotient/EEVE-Instruct-Math-10.8B") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("kuotient/EEVE-Instruct-Math-10.8B") model = AutoModelForCausalLM.from_pretrained("kuotient/EEVE-Instruct-Math-10.8B", 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=256) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use kuotient/EEVE-Instruct-Math-10.8B with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "kuotient/EEVE-Instruct-Math-10.8B" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "kuotient/EEVE-Instruct-Math-10.8B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/kuotient/EEVE-Instruct-Math-10.8B
- SGLang
How to use kuotient/EEVE-Instruct-Math-10.8B 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 "kuotient/EEVE-Instruct-Math-10.8B" \ --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": "kuotient/EEVE-Instruct-Math-10.8B", "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 "kuotient/EEVE-Instruct-Math-10.8B" \ --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": "kuotient/EEVE-Instruct-Math-10.8B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use kuotient/EEVE-Instruct-Math-10.8B with Docker Model Runner:
docker model run hf.co/kuotient/EEVE-Instruct-Math-10.8B
Update README.md
Browse files
README.md
CHANGED
|
@@ -20,12 +20,20 @@ tags:
|
|
| 20 |
- merge
|
| 21 |
|
| 22 |
---
|
| 23 |
-
# EEVE-Math-10.8B
|
| 24 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
|
| 26 |
| Model | gsm8k-ko(pass@1) |
|
| 27 |
|---|---|
|
| 28 |
-
| Base | 0.4049 |
|
| 29 |
| [EEVE-Math](https://huggingface.co/kuotient/EEVE-Math-10.8B) (epoch 1) | 0.508 |
|
| 30 |
| EEVE-Math (epoch 2) | **0.539** |
|
| 31 |
| [EEVE-Instruct](https://huggingface.co/yanolja/EEVE-Korean-Instruct-10.8B-v1.0) | 0.4511 |
|
|
@@ -38,7 +46,7 @@ This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](
|
|
| 38 |
### Models Merged
|
| 39 |
|
| 40 |
The following models were included in the merge:
|
| 41 |
-
* [kuotient/EEVE-Math-10.8B
|
| 42 |
|
| 43 |
### Configuration
|
| 44 |
|
|
@@ -48,7 +56,7 @@ The following YAML configuration was used to produce this model:
|
|
| 48 |
models:
|
| 49 |
- model: yanolja/EEVE-Korean-Instruct-10.8B-v1.0
|
| 50 |
# no parameters necessary for base model
|
| 51 |
-
- model: kuotient/EEVE-Math-10.8B
|
| 52 |
parameters:
|
| 53 |
density: 1
|
| 54 |
weight: 0.6
|
|
@@ -57,7 +65,6 @@ base_model: yanolja/EEVE-Korean-Instruct-10.8B-v1.0
|
|
| 57 |
parameters:
|
| 58 |
int8_mask: true
|
| 59 |
dtype: bfloat16
|
| 60 |
-
|
| 61 |
```
|
| 62 |
|
| 63 |
## Evaluation
|
|
|
|
| 20 |
- merge
|
| 21 |
|
| 22 |
---
|
| 23 |
+
# EEVE-Instruct-Math-10.8B
|
| 24 |
+
|
| 25 |
+
`EEVE-Math` ํ๋ก์ ํธ๋
|
| 26 |
+
- Orca-Math-200k ๋ฒ์ญ ([Orca-Math: Unlocking the potential of SLMs in Grade School Math](https://arxiv.org/pdf/2402.14830.pdf))
|
| 27 |
+
- gsm8k ๋ฒ์ญ, lm_eval ํ์ฉ
|
| 28 |
+
- Mergekit์ ์ด์ฉํ dare-ties ์ฌ์ฉ ([DARE](https://arxiv.org/abs/2311.03099))
|
| 29 |
+
|
| 30 |
+
์ ๋ํ ๋ด์ฉ์ ํฌ๊ดํ๊ณ ์์ต๋๋ค.
|
| 31 |
+
|
| 32 |
+
> ์ด ๋ชจ๋ธ์ EEVE-Math์ EEVE-Instruct์ dare-ties๋ก ๋ณํฉํ ๋ณํฉ ๋ชจ๋ธ์
๋๋ค. ์ด ํ๋ก์ ํธ๋ ์ด๋ฐ ๊ณผ์ ์ ํตํด ํนํ ๋ชจ๋ธ์ EEVE-Math์ ์ฑ๋ฅ์ ๋ง์ด ์์ง ์๊ณ Instruct ๋ชจ๋ธ์ ์ฌ์ฉ์ฑ์ ์ ์งํ ์ ์์์ ๋ณด์ฌ์ฃผ๋ Proof of concept์ ์ฑ๊ฒฉ์ ๊ฐ์ง๊ณ ์์ต๋๋ค.
|
| 33 |
|
| 34 |
| Model | gsm8k-ko(pass@1) |
|
| 35 |
|---|---|
|
| 36 |
+
| EEVE(Base) | 0.4049 |
|
| 37 |
| [EEVE-Math](https://huggingface.co/kuotient/EEVE-Math-10.8B) (epoch 1) | 0.508 |
|
| 38 |
| EEVE-Math (epoch 2) | **0.539** |
|
| 39 |
| [EEVE-Instruct](https://huggingface.co/yanolja/EEVE-Korean-Instruct-10.8B-v1.0) | 0.4511 |
|
|
|
|
| 46 |
### Models Merged
|
| 47 |
|
| 48 |
The following models were included in the merge:
|
| 49 |
+
* [kuotient/EEVE-Math-10.8B](https://huggingface.co/kuotient/EEVE-Math-10.8B)
|
| 50 |
|
| 51 |
### Configuration
|
| 52 |
|
|
|
|
| 56 |
models:
|
| 57 |
- model: yanolja/EEVE-Korean-Instruct-10.8B-v1.0
|
| 58 |
# no parameters necessary for base model
|
| 59 |
+
- model: kuotient/EEVE-Math-10.8B
|
| 60 |
parameters:
|
| 61 |
density: 1
|
| 62 |
weight: 0.6
|
|
|
|
| 65 |
parameters:
|
| 66 |
int8_mask: true
|
| 67 |
dtype: bfloat16
|
|
|
|
| 68 |
```
|
| 69 |
|
| 70 |
## Evaluation
|