Instructions to use ssmits/Falcon2-5.5B-Portuguese with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ssmits/Falcon2-5.5B-Portuguese with Transformers:

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

pipe = pipeline("text-generation", model="ssmits/Falcon2-5.5B-Portuguese", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ssmits/Falcon2-5.5B-Portuguese", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("ssmits/Falcon2-5.5B-Portuguese", trust_remote_code=True)
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

vLLM

How to use ssmits/Falcon2-5.5B-Portuguese with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ssmits/Falcon2-5.5B-Portuguese"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ssmits/Falcon2-5.5B-Portuguese",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ssmits/Falcon2-5.5B-Portuguese

SGLang

How to use ssmits/Falcon2-5.5B-Portuguese 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 "ssmits/Falcon2-5.5B-Portuguese" \
    --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": "ssmits/Falcon2-5.5B-Portuguese",
		"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 "ssmits/Falcon2-5.5B-Portuguese" \
        --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": "ssmits/Falcon2-5.5B-Portuguese",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ssmits/Falcon2-5.5B-Portuguese with Docker Model Runner:
```
docker model run hf.co/ssmits/Falcon2-5.5B-Portuguese
```

Why prune?

Even though Falcon-11B is trained on 5T tokens, it is still undertrained, as can be seen by this graph: This is why the choice is made to prune 50% of the layers. Note that ~1B of continued pre-training (~1M rows of 1k tokens) is still required to restore the perplexity of this model in the desired language. I'm planning on doing that for certain languages, depending on how much compute will be available.

sliced

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

Merge Details

Merge Method

This model was pruned using the passthrough merge method.

Models Merged

The following models were included in the merge:

tiiuae/falcon-11B

Configuration

The following YAML configuration was used to produce this model:


slices:
  - sources:
      - model: tiiuae/falcon-11B
        layer_range: [0, 24]
  - sources:
      - model: tiiuae/falcon-11B
        layer_range: [55, 59]
merge_method: passthrough
dtype: bfloat16

PruneMe has been utilized using the wikimedia/wikipedia Portuguese (pt) subset by investigating layer similarity with 2000 samples. The layer ranges for pruning were determined based on this analysis to maintain performance while reducing model size.

from transformers import AutoTokenizer, AutoModelForCausalLM
import transformers
import torch

model = "ssmits/Falcon2-5.5B-Portuguese"

tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    torch_dtype=torch.bfloat16,
)
sequences = pipeline(
   "Can you explain the concepts of Quantum Computing?",
    max_length=200,
    do_sample=True,
    top_k=10,
    num_return_sequences=1,
    eos_token_id=tokenizer.eos_token_id,
)
for seq in sequences:
    print(f"Result: {seq['generated_text']}")

💥 Falcon LLMs require PyTorch 2.0 for use with transformers!

For fast inference with Falcon, check-out Text Generation Inference! Read more in this blogpost.

Direct Use

Research on large language models; as a foundation for further specialization and finetuning for specific usecases (e.g., summarization, text generation, chatbot, etc.)

Out-of-Scope Use

Production use without adequate assessment of risks and mitigation; any use cases which may be considered irresponsible or harmful.

Bias, Risks, and Limitations

Falcon2-5.5B is trained mostly on English, but also German, Spanish, French, Italian, Portuguese, Polish, Dutch, Romanian, Czech, Swedish. It will not generalize appropriately to other languages. Furthermore, as it is trained on a large-scale corpora representative of the web, it will carry the stereotypes and biases commonly encountered online.

Recommendations

We recommend users of Falcon2-5.5B to consider finetuning it for the specific set of tasks of interest, and for guardrails and appropriate precautions to be taken for any production use.