Instructions to use RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits")
model = AutoModelForCausalLM.from_pretrained("RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits

SGLang

How to use RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits 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 "RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits" \
    --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": "RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits",
		"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 "RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits" \
        --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": "RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits with Docker Model Runner:
```
docker model run hf.co/RichardErkhov/PracticeLLM_-_SOLAR-tail-10.7B-Merge-v1.0-8bits
```

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Quantization made by Richard Erkhov.

Github

Discord

Request more models

SOLAR-tail-10.7B-Merge-v1.0 - bnb 8bits

Model creator: https://huggingface.co/PracticeLLM/
Original model: https://huggingface.co/PracticeLLM/SOLAR-tail-10.7B-Merge-v1.0/

Original model description:

language: - en - ko license: cc-by-nc-sa-4.0 pipeline_tag: text-generation model-index: - name: SOLAR-tail-10.7B-Merge-v1.0 results: - task: type: text-generation name: Text Generation dataset: name: AI2 Reasoning Challenge (25-Shot) type: ai2_arc config: ARC-Challenge split: test args: num_few_shot: 25 metrics: - type: acc_norm value: 66.13 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PracticeLLM/SOLAR-tail-10.7B-Merge-v1.0 name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: HellaSwag (10-Shot) type: hellaswag split: validation args: num_few_shot: 10 metrics: - type: acc_norm value: 86.54 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PracticeLLM/SOLAR-tail-10.7B-Merge-v1.0 name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: MMLU (5-Shot) type: cais/mmlu config: all split: test args: num_few_shot: 5 metrics: - type: acc value: 66.52 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PracticeLLM/SOLAR-tail-10.7B-Merge-v1.0 name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: TruthfulQA (0-shot) type: truthful_qa config: multiple_choice split: validation args: num_few_shot: 0 metrics: - type: mc2 value: 60.57 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PracticeLLM/SOLAR-tail-10.7B-Merge-v1.0 name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: Winogrande (5-shot) type: winogrande config: winogrande_xl split: validation args: num_few_shot: 5 metrics: - type: acc value: 84.77 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PracticeLLM/SOLAR-tail-10.7B-Merge-v1.0 name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: GSM8k (5-shot) type: gsm8k config: main split: test args: num_few_shot: 5 metrics: - type: acc value: 65.58 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=PracticeLLM/SOLAR-tail-10.7B-Merge-v1.0 name: Open LLM Leaderboard

SOLAR-tail-10.7B-Merge-v1.0

Model Details

Model Developers Kyujin Han (kyujinpy)

Method
Using Mergekit.

Merge config

slices:
  - sources:
      - model: upstage/SOLAR-10.7B-v1.0
        layer_range: [0, 48]
      - model: Yhyu13/LMCocktail-10.7B-v1
        layer_range: [0, 48]
        
merge_method: slerp
base_model: upstage/SOLAR-10.7B-v1.0

parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5 # fallback for rest of tensors
tokenizer_source: union
    
dtype: float16

Model Benchmark

Open Ko leaderboard

Follow up as Ko-link.

Model	Average	ARC	HellaSwag	MMLU	TruthfulQA	Ko-CommonGenV2
PracticeLLM/SOLAR-tail-10.7B-Merge-v1.0	48.32	45.73	56.97	38.77	38.75	61.16
jjourney1125/M-SOLAR-10.7B-v1.0	55.15	49.57	60.12	54.60	49.23	62.22

Follow up as En-link.

Model	Average	ARC	HellaSwag	MMLU	TruthfulQA	Winogrande	GSM8K
PracticeLLM/SOLAR-tail-10.7B-Merge-v1.0	71.68	66.13	86.54	66.52	60.57	84.77	65.58
kyujinpy/Sakura-SOLAR-Instruct	74.40	70.99	88.42	66.33	71.79	83.66	65.20

lm-evaluation-harness

gpt2 (pretrained=PracticeLLM/SOLAR-tail-10.7B-Merge-v1.0), limit: None, provide_description: False, num_fewshot: 0, batch_size: None
|      Task      |Version| Metric |Value |   |Stderr|
|----------------|------:|--------|-----:|---|-----:|
|kobest_boolq    |      0|acc     |0.5021|±  |0.0133|
|                |       |macro_f1|0.3343|±  |0.0059|
|kobest_copa     |      0|acc     |0.6220|±  |0.0153|
|                |       |macro_f1|0.6217|±  |0.0154|
|kobest_hellaswag|      0|acc     |0.4380|±  |0.0222|
|                |       |acc_norm|0.5380|±  |0.0223|
|                |       |macro_f1|0.4366|±  |0.0222|
|kobest_sentineg |      0|acc     |0.4962|±  |0.0251|
|                |       |macro_f1|0.3316|±  |0.0113|

Implementation Code

### KO-Platypus
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

repo = "PracticeLLM/SOLAR-tail-10.7B-Merge-v1.0"
OpenOrca = AutoModelForCausalLM.from_pretrained(
        repo,
        return_dict=True,
        torch_dtype=torch.float16,
        device_map='auto'
)
OpenOrca_tokenizer = AutoTokenizer.from_pretrained(repo)

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	71.68
AI2 Reasoning Challenge (25-Shot)	66.13
HellaSwag (10-Shot)	86.54
MMLU (5-Shot)	66.52
TruthfulQA (0-shot)	60.57
Winogrande (5-shot)	84.77
GSM8k (5-shot)	65.58

Downloads last month: 4

Safetensors

Model size

11B params

Tensor type

F32

F16