Instructions to use c-bone/CrystaLLM-pi_mp_20_base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use c-bone/CrystaLLM-pi_mp_20_base with Transformers:

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

pipe = pipeline("text-generation", model="c-bone/CrystaLLM-pi_mp_20_base")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("c-bone/CrystaLLM-pi_mp_20_base")
model = AutoModelForCausalLM.from_pretrained("c-bone/CrystaLLM-pi_mp_20_base", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use c-bone/CrystaLLM-pi_mp_20_base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "c-bone/CrystaLLM-pi_mp_20_base"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "c-bone/CrystaLLM-pi_mp_20_base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/c-bone/CrystaLLM-pi_mp_20_base

SGLang

How to use c-bone/CrystaLLM-pi_mp_20_base 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 "c-bone/CrystaLLM-pi_mp_20_base" \
    --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": "c-bone/CrystaLLM-pi_mp_20_base",
		"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 "c-bone/CrystaLLM-pi_mp_20_base" \
        --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": "c-bone/CrystaLLM-pi_mp_20_base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use c-bone/CrystaLLM-pi_mp_20_base with Docker Model Runner:
```
docker model run hf.co/c-bone/CrystaLLM-pi_mp_20_base
```

CrystaLLM-pi_mp_20_base / training_args.json

c-bone

Use the CrystaLLM-pi training config: public c-bone/mp_20_pxrd (same CIFs as the training data), muon_lr, adam_beta2 and min_lr_rate read from the checkpoint

d112d0e verified 16 days ago

Raw History Blame Contribute Delete

2.11 kB

	{
	"activate_conditionality": null,
	"adam_beta1": 0.9,
	"adam_beta2": 0.995,
	"attention_dropout": 0.1,
	"codecarbon": true,
	"cond_dropout": null,
	"cond_lr": null,
	"cond_wd": null,
	"condition_columns": null,
	"config": "_config_files/training/unconditional/mp-20-text.jsonc",
	"context_extension_warmup_steps": 0,
	"context_length": 1024,
	"data_seed": 1,
	"dataset_HF": "c-bone/mp_20_pxrd",
	"deepspeed_config": "_config_files/deepspeed_default.json",
	"do_sample": "True",
	"early_stopping_patience": 15,
	"early_stopping_threshold": 5e-06,
	"embedding_dropout": 0.1,
	"eval_batch_size": 32,
	"eval_steps": 1000,
	"eval_strategy": "steps",
	"fp16": false,
	"gen_max_length": 1024,
	"grad_clip": 1.0,
	"gradient_accumulation_steps": 1,
	"greater_is_better": false,
	"input_parquet": null,
	"learning_rate": 0.002,
	"load_best_model_at_end": true,
	"logging_steps": 100,
	"lr_scheduler_kwargs": {
	"min_lr_rate": 0.01
	},
	"lr_scheduler_type": "cosine_with_min_lr",
	"max_return_attempts": 1,
	"max_samples": null,
	"max_steps": 5000,
	"metric_for_best_model": "eval_loss",
	"model_ckpt_dir": "model_ckpts/cif-gpt2-small/checkpoint-400",
	"muon_lr": 0.02,
	"muon_momentum": 0.95,
	"n_embd": 512,
	"n_head": 8,
	"n_heads_sharing_slider": null,
	"n_hidden_cond": null,
	"n_layer": 8,
	"n_prefix_tokens": null,
	"num_return_sequences": 1,
	"optimizer": "muon",
	"output_dir": "model_ckpts/mp_20_text/",
	"output_parquet": null,
	"pretrained_model_dir": null,
	"pretrained_tokenizer_dir": "HF-cif-tokenizer",
	"remove_CIFs_above_context": false,
	"remove_CIFs_with_unk": false,
	"report_to": "wandb",
	"residual_dropout": 0.1,
	"save_strategy": "steps",
	"save_total_limit": 2,
	"scoring_mode": "None",
	"screening_profile": "application",
	"seed": 1,
	"target_valid_cifs": 1,
	"temperature": 1.0,
	"top_k": 15,
	"top_p": 0.95,
	"torch_compile": true,
	"tracker_project": "CrystaLLM-pi",
	"train_batch_size": 32,
	"wandb_project_folder": "CrystaLLM-lematbench",
	"warmup_ratio": 0.0,
	"warmup_steps": null,
	"weight_decay": 0.1
	}