Falcon2-11B is an 11B parameters causal decoder-only model built by TII and trained on over 5,000B tokens of RefinedWeb enhanced with curated corpora. The model is made available under the TII Falcon License 2.0, the permissive Apache 2.0-based software license which includes an acceptable use policy that promotes the responsible use of AI.
Paper coming soon ๐.
๐ค To get started with Falcon (inference, finetuning, quantization, etc.), we recommend reading this great blogpost from HF!
โ ๏ธ This is a raw, pretrained model, which should be further finetuned for most usecases.
from transformers import AutoTokenizer, AutoModelForCausalLM
import transformers
import torch
model = "tiiuae/falcon-11B"
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']}")
๐ฅ 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.
Model Card for Falcon2-11B Model Details Model DescriptionResearch 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 UseProduction use without adequate assessment of risks and mitigation; any use cases which may be considered irresponsible or harmful.
Bias, Risks, and LimitationsFalcon2-11B 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.
RecommendationsWe recommend users of Falcon2-11B 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.
How to Get Started with the Modelfrom transformers import AutoTokenizer, AutoModelForCausalLM
import transformers
import torch
model = "tiiuae/falcon-11B"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
torch_dtype=torch.bfloat16,
device_map="auto",
)
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']}")
Falcon2-11B was trained over 5,000B tokens of RefinedWeb, a high-quality filtered and deduplicated web dataset which we enhanced with curated corpora. It followed a four stage training strategy. The first three stages were focused on increasing the context length, from to 2048 to 4096 and finally to 8192 tokens. The last stage aimed to further enhance performance using only high quality data.
Overall, the data sources included RefinedWeb-English, Refined Web-Europe (cs, de, es, fr, it, nl, pl, pt, ro, sv), high quality technical data, code data, and conversational data extracted from public sources.
The training stages were as follows:
Stage Context length Tokens Stage 1 2048 4500 B Stage 2 4096 250 B Stage 3 8192 250 B Stage 4 8192 500 BThe data was tokenized with the Falcon-7B/11B tokenizer.
Training ProcedureFalcon2-11B was trained on 1024 A100 40GB GPUs for the majority of the training, using a 3D parallelism strategy (TP=8, PP=1, DP=128) combined with ZeRO and Flash-Attention 2.
Training Hyperparameters Hyperparameter Value Comment Precisionbfloat16 Optimizer AdamW Max learning rate 3.7e-4 Following a linear warm-up, then cosine decay to 1.89e-5 across 4500 B tokens. Weight decay 1e-1 Z-loss 1e-4 Batch size Variable Batch size was gradually increased during the training Speeds, Sizes, Times
The model training took roughly two months.
Evaluation English Benchmark Value ARC-Challenge-25shots 59.73 HellaSwag-10shots 82.91 MMLU-5shots 58.37 Winogrande-5shots 78.30 TruthfulQA-0shot 52.56 GSM8k-5shots 53.83 ARC-Challenge-0shot 50.17 ARC-Easy-0shot 77.78 Hellaswag-0shot 82.07We thank the leaderboard team from HuggingFace for providing an official evaluation of our model on the leaderboard tasks.
Technical Specifications Model Architecture and ObjectiveFalcon2-11B is a causal decoder-only model trained on a causal language modeling task (i.e., predict the next token).
The architecture is broadly adapted from the GPT-3 paper (Brown et al., 2020), with the following differences:
d_model 4096 head_dim 128 Vocabulary 65024 Sequence length 8192 During stages 3 and 4 Compute Infrastructure Hardware
Falcon2-11B was trained on AWS SageMaker, using on average 1024 A100 40GB GPUs in 128 p4d instances.
SoftwareFalcon2-11B was trained a custom distributed training codebase, Gigatron. It uses a 3D parallelism approach combined with ZeRO, high-performance Triton kernels and FlashAttention-2. More details about the distributed training strategy can be found in Almazrouei et.al.
CitationPaper coming soon ๐.
LicenseFalcon2-11B is licenced under TII Falcon License 2.0, the permissive Apache 2.0-based software license which includes an acceptable use policy that promotes the responsible use of AI.
ContactRetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4