Instructions to use CLMBR/npi-only-transformer-3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use CLMBR/npi-only-transformer-3 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="CLMBR/npi-only-transformer-3")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("CLMBR/npi-only-transformer-3") model = AutoModelForCausalLM.from_pretrained("CLMBR/npi-only-transformer-3") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use CLMBR/npi-only-transformer-3 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "CLMBR/npi-only-transformer-3" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "CLMBR/npi-only-transformer-3", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/CLMBR/npi-only-transformer-3
- SGLang
How to use CLMBR/npi-only-transformer-3 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 "CLMBR/npi-only-transformer-3" \ --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": "CLMBR/npi-only-transformer-3", "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 "CLMBR/npi-only-transformer-3" \ --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": "CLMBR/npi-only-transformer-3", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use CLMBR/npi-only-transformer-3 with Docker Model Runner:
docker model run hf.co/CLMBR/npi-only-transformer-3
metadata
tags:
- generated_from_trainer
model-index:
- name: npi-only-transformer-3
results: []
npi-only-transformer-3
This model is a fine-tuned version of on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.8598
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 3
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 3052726
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.2223 | 0.03 | 76320 | 4.1964 |
| 4.0204 | 1.03 | 152640 | 4.0268 |
| 3.912 | 0.03 | 228960 | 3.9523 |
| 3.8408 | 1.03 | 305280 | 3.9111 |
| 3.7917 | 0.03 | 381600 | 3.8861 |
| 3.7492 | 1.03 | 457920 | 3.8700 |
| 3.7159 | 0.03 | 534240 | 3.8608 |
| 3.6895 | 1.03 | 610560 | 3.8526 |
| 3.6619 | 0.03 | 686880 | 3.8481 |
| 3.6343 | 1.03 | 763200 | 3.8460 |
| 3.61 | 0.03 | 839520 | 3.8443 |
| 3.5902 | 1.03 | 915840 | 3.8437 |
| 3.571 | 0.03 | 992160 | 3.8429 |
| 3.5525 | 1.03 | 1068480 | 3.8434 |
| 3.5337 | 0.03 | 1144800 | 3.8455 |
| 3.5324 | 1.03 | 1221120 | 3.8451 |
| 3.5107 | 0.03 | 1297440 | 3.8464 |
| 3.4996 | 1.03 | 1373760 | 3.8468 |
| 3.4875 | 0.03 | 1450080 | 3.8484 |
| 3.475 | 1.03 | 1526400 | 3.8496 |
| 3.4666 | 0.03 | 1602720 | 3.8495 |
| 3.4571 | 1.03 | 1679040 | 3.8516 |
| 3.4483 | 0.03 | 1755360 | 3.8525 |
| 3.4417 | 1.03 | 1831680 | 3.8534 |
| 3.4295 | 0.03 | 1908000 | 3.8552 |
| 3.4152 | 1.03 | 1984320 | 3.8558 |
| 3.3995 | 0.03 | 2060640 | 3.8572 |
| 3.3901 | 1.03 | 2136960 | 3.8578 |
| 3.3801 | 0.03 | 2213280 | 3.8582 |
| 3.367 | 1.03 | 2289600 | 3.8592 |
| 3.3558 | 0.03 | 2365920 | 3.8611 |
| 3.3561 | 1.03 | 2442240 | 3.8599 |
| 3.3408 | 0.03 | 2518560 | 3.8615 |
| 3.334 | 1.03 | 2594880 | 3.8621 |
| 3.3245 | 0.03 | 2671200 | 3.8619 |
| 3.317 | 0.03 | 2747520 | 3.8619 |
| 3.3107 | 1.03 | 2823840 | 3.8615 |
| 3.3063 | 0.03 | 2900160 | 3.8617 |
| 3.3022 | 1.03 | 2976480 | 3.8610 |
| 3.2972 | 0.02 | 3052726 | 3.8598 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.12.0
- Tokenizers 0.13.3