Instructions to use PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B with Transformers:

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

pipe = pipeline("text-generation", model="PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B")
model = AutoModelForCausalLM.from_pretrained("PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B

SGLang

How to use PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B 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 "PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B" \
    --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": "PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B",
		"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 "PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B" \
        --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": "PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B with Docker Model Runner:
```
docker model run hf.co/PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

merge

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

Merge Details

Merge Method

This model was merged using the Passthrough merge method.

Models Merged

The following models were included in the merge:

mistralai/Mistral-Nemo-Base-2407

Configuration

The following YAML configuration was used to produce this model:

merge_method: passthrough
dtype: bfloat16
slices:
# Original first 10 layers (L0-L9)
- sources:
  - layer_range: [0, 10]
    model: mistralai/Mistral-Nemo-Base-2407

# "Before Dupe" of L10
- sources:
  - layer_range: [10, 11]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
# Original L10
- sources:
  - layer_range: [10, 11]
    model: mistralai/Mistral-Nemo-Base-2407

# "Before Dupe" of L11
- sources:
  - layer_range: [11, 12]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
# Original L11
- sources:
  - layer_range: [11, 12]
    model: mistralai/Mistral-Nemo-Base-2407

# "Before Dupe" of L12
- sources:
  - layer_range: [12, 13]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
# Original L12
- sources:
  - layer_range: [12, 13]
    model: mistralai/Mistral-Nemo-Base-2407

# "Before Dupe" of L13
- sources:
  - layer_range: [13, 14]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
# Original L13
- sources:
  - layer_range: [13, 14]
    model: mistralai/Mistral-Nemo-Base-2407

# "Before Dupe" of L14
- sources:
  - layer_range: [14, 15]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
# Original L14
- sources:
  - layer_range: [14, 15]
    model: mistralai/Mistral-Nemo-Base-2407

# "Before Dupe" of L15
- sources:
  - layer_range: [15, 16]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
# Original L15
- sources:
  - layer_range: [15, 16]
    model: mistralai/Mistral-Nemo-Base-2407

# "Before Dupe" of L16
- sources:
  - layer_range: [16, 17]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
# Original L16
- sources:
  - layer_range: [16, 17]
    model: mistralai/Mistral-Nemo-Base-2407

# "Before Dupe" of L17
- sources:
  - layer_range: [17, 18]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
# Original L17
- sources:
  - layer_range: [17, 18]
    model: mistralai/Mistral-Nemo-Base-2407

# "Before Dupe" of L18
- sources:
  - layer_range: [18, 19]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
# Original L18
- sources:
  - layer_range: [18, 19]
    model: mistralai/Mistral-Nemo-Base-2407

# "Before Dupe" of L19
- sources:
  - layer_range: [19, 20]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0
# Original L19
- sources:
  - layer_range: [19, 20]
    model: mistralai/Mistral-Nemo-Base-2407

# Original L20
- sources:
  - layer_range: [20, 21]
    model: mistralai/Mistral-Nemo-Base-2407
# "After Dupe" of L20
- sources:
  - layer_range: [20, 21]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0

# Original L21
- sources:
  - layer_range: [21, 22]
    model: mistralai/Mistral-Nemo-Base-2407
# "After Dupe" of L21
- sources:
  - layer_range: [21, 22]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0

# Original L22
- sources:
  - layer_range: [22, 23]
    model: mistralai/Mistral-Nemo-Base-2407
# "After Dupe" of L22
- sources:
  - layer_range: [22, 23]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0

# Original L23
- sources:
  - layer_range: [23, 24]
    model: mistralai/Mistral-Nemo-Base-2407
# "After Dupe" of L23
- sources:
  - layer_range: [23, 24]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0

# Original L24
- sources:
  - layer_range: [24, 25]
    model: mistralai/Mistral-Nemo-Base-2407
# "After Dupe" of L24
- sources:
  - layer_range: [24, 25]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0

# Original L25
- sources:
  - layer_range: [25, 26]
    model: mistralai/Mistral-Nemo-Base-2407
# "After Dupe" of L25
- sources:
  - layer_range: [25, 26]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0

# Original L26
- sources:
  - layer_range: [26, 27]
    model: mistralai/Mistral-Nemo-Base-2407
# "After Dupe" of L26
- sources:
  - layer_range: [26, 27]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0

# Original L27
- sources:
  - layer_range: [27, 28]
    model: mistralai/Mistral-Nemo-Base-2407
# "After Dupe" of L27
- sources:
  - layer_range: [27, 28]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0

# Original L28
- sources:
  - layer_range: [28, 29]
    model: mistralai/Mistral-Nemo-Base-2407
# "After Dupe" of L28
- sources:
  - layer_range: [28, 29]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0

# Original L29
- sources:
  - layer_range: [29, 30]
    model: mistralai/Mistral-Nemo-Base-2407
# "After Dupe" of L29
- sources:
  - layer_range: [29, 30]
    model: mistralai/Mistral-Nemo-Base-2407
    parameters:
      scale:
      - filter: o_proj
        value: 0.0
      - filter: down_proj
        value: 0.0
      - value: 1.0

# Original last 10 layers (L30-L39)
- sources:
  - layer_range: [30, 40]
    model: mistralai/Mistral-Nemo-Base-2407

Downloads last month: -

Safetensors

Model size

18B params

Tensor type

BF16

Model tree for PJMixers-Dev/MN-2407-Base-ChatML-Zeroed-Interleaved-17.7B

Base model

mistralai/Mistral-Nemo-Base-2407

Finetuned

(88)

this model

Quantizations

2 models