Instructions to use top-50000/testing-2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use top-50000/testing-2 with Transformers:

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

pipe = pipeline("text-generation", model="top-50000/testing-2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("top-50000/testing-2")
model = AutoModelForCausalLM.from_pretrained("top-50000/testing-2")
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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use top-50000/testing-2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "top-50000/testing-2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "top-50000/testing-2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/top-50000/testing-2

SGLang

How to use top-50000/testing-2 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 "top-50000/testing-2" \
    --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": "top-50000/testing-2",
		"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 "top-50000/testing-2" \
        --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": "top-50000/testing-2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use top-50000/testing-2 with Docker Model Runner:
```
docker model run hf.co/top-50000/testing-2
```

merged_4b

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the TIES merge method using ./evolve_storage/input_models/Qwen3-4B_131373129 as a base.

Models Merged

The following models were included in the merge:

./evolve_storage/input_models/4b_1_2932744254
./evolve_storage/input_models/4b_2_935817604

Configuration

The following YAML configuration was used to produce this model:

base_model: ./evolve_storage/input_models/Qwen3-4B_131373129
dtype: bfloat16
merge_method: ties
parameters:
  int8_mask: 1.0
  normalize: 1.0
slices:
- sources:
  - layer_range: [0, 6]
    model: ./evolve_storage/input_models/4b_1_2932744254
    parameters:
      density:
      - filter: mlp
        value: 1.0
      - filter: self_attn
        value: 1.0
      - value: 0.8608594704318231
      weight:
      - filter: mlp
        value: 0.3490062747244206
      - filter: self_attn
        value: 0.09060462688545917
      - value: 0.7576223803340598
  - layer_range: [0, 6]
    model: ./evolve_storage/input_models/4b_2_935817604
    parameters:
      density:
      - filter: mlp
        value: 0.9307034739117228
      - filter: self_attn
        value: 0.7266676131372826
      - value: 0.7992834608507406
      weight:
      - filter: mlp
        value: 0.02382702396781508
      - filter: self_attn
        value: 0.19373265930727943
      - value: 0.5095495553391711
  - layer_range: [0, 6]
    model: ./evolve_storage/input_models/Qwen3-4B_131373129
- sources:
  - layer_range: [6, 12]
    model: ./evolve_storage/input_models/4b_1_2932744254
    parameters:
      density:
      - filter: mlp
        value: 1.0
      - filter: self_attn
        value: 0.9956934876179888
      - value: 1.0
      weight:
      - filter: mlp
        value: 0.7637369917121993
      - filter: self_attn
        value: 0.7579899664980693
      - value: 0.369874736821347
  - layer_range: [6, 12]
    model: ./evolve_storage/input_models/4b_2_935817604
    parameters:
      density:
      - filter: mlp
        value: 1.0
      - filter: self_attn
        value: 0.7931662945197395
      - value: 1.0
      weight:
      - filter: mlp
        value: 0.05065405710236617
      - filter: self_attn
        value: 0.1681017405390699
      - value: 0.278174377018312
  - layer_range: [6, 12]
    model: ./evolve_storage/input_models/Qwen3-4B_131373129
- sources:
  - layer_range: [12, 18]
    model: ./evolve_storage/input_models/4b_1_2932744254
    parameters:
      density:
      - filter: mlp
        value: 1.0
      - filter: self_attn
        value: 1.0
      - value: 1.0
      weight:
      - filter: mlp
        value: 0.1896079785687373
      - filter: self_attn
        value: 0.28109871222289246
      - value: 0.5919682098365076
  - layer_range: [12, 18]
    model: ./evolve_storage/input_models/4b_2_935817604
    parameters:
      density:
      - filter: mlp
        value: 0.9657611673995403
      - filter: self_attn
        value: 0.7203176611210709
      - value: 0.5429222666986293
      weight:
      - filter: mlp
        value: 0.6105068874861986
      - filter: self_attn
        value: 0.35883124159464513
      - value: 0.08904009475761779
  - layer_range: [12, 18]
    model: ./evolve_storage/input_models/Qwen3-4B_131373129
- sources:
  - layer_range: [18, 24]
    model: ./evolve_storage/input_models/4b_1_2932744254
    parameters:
      density:
      - filter: mlp
        value: 0.9149303352862976
      - filter: self_attn
        value: 1.0
      - value: 0.877602461891939
      weight:
      - filter: mlp
        value: 0.4686086798952363
      - filter: self_attn
        value: 0.9038594427840757
      - value: 0.10082926037736678
  - layer_range: [18, 24]
    model: ./evolve_storage/input_models/4b_2_935817604
    parameters:
      density:
      - filter: mlp
        value: 0.7101112464063535
      - filter: self_attn
        value: 0.7817731468186313
      - value: 0.39642301784045736
      weight:
      - filter: mlp
        value: 0.12881047196006434
      - filter: self_attn
        value: 0.5913495175648413
      - value: 0.4608092365074321
  - layer_range: [18, 24]
    model: ./evolve_storage/input_models/Qwen3-4B_131373129
- sources:
  - layer_range: [24, 30]
    model: ./evolve_storage/input_models/4b_1_2932744254
    parameters:
      density:
      - filter: mlp
        value: 0.5207504504970493
      - filter: self_attn
        value: 0.7016905835163534
      - value: 0.7841906098086426
      weight:
      - filter: mlp
        value: 0.03936529169370809
      - filter: self_attn
        value: 0.3540197424712478
      - value: 0.23794897877129362
  - layer_range: [24, 30]
    model: ./evolve_storage/input_models/4b_2_935817604
    parameters:
      density:
      - filter: mlp
        value: 0.6658594462559264
      - filter: self_attn
        value: 1.0
      - value: 0.7557856184584113
      weight:
      - filter: mlp
        value: 0.5961021443524465
      - filter: self_attn
        value: 0.5159043688030769
      - value: 0.7605556292871718
  - layer_range: [24, 30]
    model: ./evolve_storage/input_models/Qwen3-4B_131373129
- sources:
  - layer_range: [30, 36]
    model: ./evolve_storage/input_models/4b_1_2932744254
    parameters:
      density:
      - filter: mlp
        value: 1.0
      - filter: self_attn
        value: 1.0
      - value: 0.63120667056331
      weight:
      - filter: mlp
        value: 0.15069904386157829
      - filter: self_attn
        value: 0.6308542464264801
      - value: 0.4804838636683506
  - layer_range: [30, 36]
    model: ./evolve_storage/input_models/4b_2_935817604
    parameters:
      density:
      - filter: mlp
        value: 1.0
      - filter: self_attn
        value: 1.0
      - value: 0.924178196118008
      weight:
      - filter: mlp
        value: 0.5716878395339086
      - filter: self_attn
        value: 0.19598590068808175
      - value: 0.8797652412255632
  - layer_range: [30, 36]
    model: ./evolve_storage/input_models/Qwen3-4B_131373129

Downloads last month: 22

Safetensors

Model size

4B params

Tensor type

BF16

Paper for top-50000/testing-2

Resolving Interference When Merging Models

Paper • 2306.01708 • Published Jun 2, 2023 • 19