Instructions to use Defetya/math-kaggle-comp with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Defetya/math-kaggle-comp with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Defetya/math-kaggle-comp", trust_remote_code=True)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("Defetya/math-kaggle-comp", trust_remote_code=True) model = AutoModelForCausalLM.from_pretrained("Defetya/math-kaggle-comp", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Defetya/math-kaggle-comp with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Defetya/math-kaggle-comp" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Defetya/math-kaggle-comp", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Defetya/math-kaggle-comp
- SGLang
How to use Defetya/math-kaggle-comp 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 "Defetya/math-kaggle-comp" \ --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": "Defetya/math-kaggle-comp", "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 "Defetya/math-kaggle-comp" \ --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": "Defetya/math-kaggle-comp", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use Defetya/math-kaggle-comp with Docker Model Runner:
docker model run hf.co/Defetya/math-kaggle-comp
| { | |
| "_name_or_path": "microsoft/phi-2", | |
| "architectures": [ | |
| "PhiForCausalLM" | |
| ], | |
| "attention_dropout": 0.0, | |
| "attention_partition_spec": [ | |
| [ | |
| "dp", | |
| "fsdp" | |
| ], | |
| "sp", | |
| "tp", | |
| null | |
| ], | |
| "attn_mechanism": "normal", | |
| "auto_map": { | |
| "AutoConfig": "microsoft/phi-2--configuration_phi.PhiConfig", | |
| "AutoModelForCausalLM": "microsoft/phi-2--modeling_phi.PhiForCausalLM" | |
| }, | |
| "axis_dims": [ | |
| 1, | |
| -1, | |
| 1, | |
| 1 | |
| ], | |
| "axis_names": [ | |
| "dp", | |
| "fsdp", | |
| "tp", | |
| "sp" | |
| ], | |
| "backend": null, | |
| "bias_partition_spec": [ | |
| [ | |
| "dp", | |
| "fsdp" | |
| ], | |
| null, | |
| null, | |
| null | |
| ], | |
| "bits": null, | |
| "block_b": 1, | |
| "block_k": 128, | |
| "block_k_dkv": 128, | |
| "block_k_dq": 128, | |
| "block_k_major": 128, | |
| "block_k_major_dkv": 128, | |
| "block_k_major_dq": 128, | |
| "block_q": 128, | |
| "block_q_dkv": 128, | |
| "block_q_dq": 128, | |
| "block_q_major_dkv": 128, | |
| "bos_token_id": 50256, | |
| "easy_method": "train", | |
| "embd_pdrop": 0.0, | |
| "eos_token_id": 50256, | |
| "generation_attention_partition_spec": [ | |
| [ | |
| "dp", | |
| "fsdp" | |
| ], | |
| null, | |
| "tp", | |
| null | |
| ], | |
| "generation_bias_partition_spec": [ | |
| [ | |
| "dp", | |
| "fsdp" | |
| ], | |
| null, | |
| null, | |
| null | |
| ], | |
| "generation_query_partition_spec": [ | |
| [ | |
| "dp", | |
| "fsdp" | |
| ], | |
| "sp", | |
| null, | |
| null | |
| ], | |
| "hidden_act": "gelu_new", | |
| "initializer_range": 0.02, | |
| "intermediate_size": 10240, | |
| "key_partition_spec": [ | |
| [ | |
| "dp", | |
| "fsdp" | |
| ], | |
| "sp", | |
| "tp", | |
| null | |
| ], | |
| "layer_norm_eps": 1e-05, | |
| "model_type": "phi", | |
| "n_embd": 2560, | |
| "n_positions": 2048, | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 32, | |
| "num_key_value_heads": 32, | |
| "partial_rotary_factor": 0.4, | |
| "qk_layernorm": false, | |
| "query_partition_spec": [ | |
| [ | |
| "dp", | |
| "fsdp" | |
| ], | |
| "sp", | |
| "tp", | |
| null | |
| ], | |
| "resid_pdrop": 0.1, | |
| "rope_scaling": null, | |
| "rope_theta": 10000.0, | |
| "scan_attention_layers": false, | |
| "scan_mlp_chunk_size": 1024, | |
| "scan_ring_attention": false, | |
| "shard_attention_computation": true, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.38.0", | |
| "use_cache": true, | |
| "use_scan_mlp": true, | |
| "use_sharded_kv_caching": true, | |
| "use_sharding_constraint": false, | |
| "value_partition_spec": [ | |
| [ | |
| "dp", | |
| "fsdp" | |
| ], | |
| "sp", | |
| "tp", | |
| null | |
| ], | |
| "vocab_size": 51200 | |
| } | |