metadata

language:
  - en
tags:
  - falcon3
  - falcon3_mamba
base_model:
  - tiiuae/Falcon3-Mamba-7B-Base

Falcon3-Mamba-7B-Instruct

Falcon3 family of Open Foundation Models is a set of pretrained and instruct LLMs ranging from 1B to 10B.

This repository contains the Falcon3-Mamba-7B-Instruct. It achieves ,compared to similar SSM-based models of the same size, state of art results (at release's time) on reasoning, language understanding, instruction following, code and mathematics tasks. Falcon3-Mamba-7B-Instruct supports a context length up to 32K and 1 language (english).

Model Details

Architecture(same as Falcon-Mamba-7b)
- Mamba1 based causal decoder only architecture trained on a causal language modeling task (i.e., predict the next token).
- 64 decoder blocks
- width: 4096
- state_size: 16
- 32k context length
- 65k vocab size
Pretrained on 7 Teratokens of datasets comprising of web, code, STEM and high quality data using 2048 H100 GPU chips
Postrained on 1.2 million samples of STEM, conversations, code, and safety.
Developed by Technology Innovation Institute
License: TII Falcon-LLM License 2.0
Model Release Date: December 2024

Getting started

Click to expand

from transformers import AutoTokenizer, AutoModelForCausalLM


from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "tiiuae/Falcon3-Mamba-7B-Instruct"

model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)

prompt = "How many hours in one day?"
messages = [
    {"role": "system", "content": "You are a helpful friendly assistant Falcon3 from TII, try to follow instructions as much as possible."},
    {"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=1024
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
print(response)

Benchmarks

We report in the following table our internal pipeline benchmarks:

Category	Benchmark	Zamba2-7B-instruct	Jamba-1.5-Mini-instruct	falcon-mamba-7b-instruct	Llama-3.1-8B-Instruct	Qwen2-7B-Instruct
General	MMLU (5-shot)	-	-	-	-	-
	MMLU-PRO (5-shot)	-	-	-	-	-
	IFEval	-	-	-	-	-
Math	GSM8K (5-shot)	-	-	-	-	-
Math	MATH(4-shot)	-	-	-	-	-
Reasoning	Arc Challenge (25-shot)	-	-	-	-	-
	GPQA (0-shot)	-	-	-	-	-
	MUSR (0-shot)	-	-	-	-	-
	BBH (3-shot)	-	-	-	-	-
CommonSense Understanding	PIQA (0-shot)	-	-	-	-	-
	SciQ (0-shot)	-	-	-	-	-
	Winogrande (0-shot)	-	-	-	-	-
	OpenbookQA (0-shot)	-	-	-	-	-

Citation

If Falcon3 family were helpful to your work, feel free to give us a cite.

@misc{Falcon3,
    title = {The Falcon 3 family of Open Models},
    author = {TII Team},
    month = {December},
    year = {2024}
}