Instructions to use caikybaldo999/CBLM-Code-300M-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use caikybaldo999/CBLM-Code-300M-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="caikybaldo999/CBLM-Code-300M-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("caikybaldo999/CBLM-Code-300M-Instruct")
model = AutoModelForCausalLM.from_pretrained("caikybaldo999/CBLM-Code-300M-Instruct", device_map="auto")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use caikybaldo999/CBLM-Code-300M-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "caikybaldo999/CBLM-Code-300M-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "caikybaldo999/CBLM-Code-300M-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/caikybaldo999/CBLM-Code-300M-Instruct

SGLang

How to use caikybaldo999/CBLM-Code-300M-Instruct 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 "caikybaldo999/CBLM-Code-300M-Instruct" \
    --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": "caikybaldo999/CBLM-Code-300M-Instruct",
		"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 "caikybaldo999/CBLM-Code-300M-Instruct" \
        --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": "caikybaldo999/CBLM-Code-300M-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use caikybaldo999/CBLM-Code-300M-Instruct with Docker Model Runner:
```
docker model run hf.co/caikybaldo999/CBLM-Code-300M-Instruct
```

caikybaldo999 commited on Jul 20

Commit

ed204ef

verified ·

1 Parent(s): 9db68f8

CBLM Code 300M instruction tuning

Browse files

Files changed (7) hide show

README.md +33 -0
chat_template.jinja +10 -0
config.json +32 -0
generation_config.json +11 -0
model.safetensors +3 -0
tokenizer.json +0 -0
tokenizer_config.json +20 -0

README.md ADDED Viewed

	@@ -0,0 +1,33 @@

+---
+library_name: transformers
+pipeline_tag: text-generation
+language:
+- pt
+- en
+base_model: caikybaldo999/CBLM-Code-300M
+tags:
+- llama
+- decoder-only
+- instruct
+- code
+- portuguese
+---
+# CBLM Code 300M Instruct
+Versão instruction-tuned de [caikybaldo999/CBLM-Code-300M](https://huggingface.co/caikybaldo999/CBLM-Code-300M).
+- Arquitetura preservada do modelo-base: `LlamaForCausalLM`
+- Parâmetros: `303,350,784`
+- Contexto de SFT: `1024` tokens
+- Treinamento completo dos pesos em BF16/FP16
+- Dados de instrução em português e programação
+## Formato de chat
+Use `tokenizer.apply_chat_template(messages, add_generation_prompt=True)`.
+## Limitações
+Modelo experimental de aproximadamente 300M parâmetros. Pode errar fatos,
+produzir código incorreto e gerar conteúdo inesperado. Verifique as respostas.

chat_template.jinja ADDED Viewed

	@@ -0,0 +1,10 @@

+{% for message in messages %}{% if message['role'] == 'system' %}### Sistema:
+{{ message['content'] }}
+{% elif message['role'] == 'user' %}### Usuário:
+{{ message['content'] }}
+{% elif message['role'] == 'assistant' %}### Assistente:
+{{ message['content'] }}{{ eos_token }}
+{% endif %}{% endfor %}{% if add_generation_prompt %}### Assistente:
+{% endif %}

config.json ADDED Viewed

	@@ -0,0 +1,32 @@

+{
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "dtype": "bfloat16",
+  "eos_token_id": 2,
+  "head_dim": 64,
+  "hidden_act": "silu",
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 2304,
+  "max_position_embeddings": 1024,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 16,
+  "pad_token_id": 0,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_parameters": {
+    "rope_theta": 10000.0,
+    "rope_type": "default"
+  },
+  "tie_word_embeddings": true,
+  "transformers_version": "5.14.1",
+  "use_cache": true,
+  "vocab_size": 32000
+}

generation_config.json ADDED Viewed

	@@ -0,0 +1,11 @@

+{
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": 2,
+  "max_new_tokens": 256,
+  "pad_token_id": 0,
+  "repetition_penalty": 1.1,
+  "temperature": 0.8,
+  "top_p": 0.95,
+  "transformers_version": "5.14.1"
+}

model.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:22febd2130ee0e387ed0e396acd9c56571bc5017e5cc488d04590983f4764f0d
+size 606726400

tokenizer.json ADDED Viewed

The diff for this file is too large to render. See raw diff

tokenizer_config.json ADDED Viewed

	@@ -0,0 +1,20 @@

+{
+  "backend": "tokenizers",
+  "bos_token": "<|bos|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|eos|>",
+  "is_local": false,
+  "local_files_only": false,
+  "max_length": 1024,
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1024,
+  "pad_token": "<|pad|>",
+  "return_token_type_ids": false,
+  "stride": 0,
+  "tokenizer_class": "TokenizersBackend",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first"
+}