Finally, the model we started training with our blog post is ready 🎉
The model might contain some problems, especially with the FIM format
It Works As a Chat
The primary application of this model is code completion (infill) in multiple programming languages.
But it works as a chat quite well.
Example
Fill-in-the-middle uses special tokens to identify the prefix/middle/suffix part of the input and output:
python
1# pip install -q transformers2from transformers import AutoModelForCausalLM, AutoTokenizer
34checkpoint ="smallcloudai/Refact-1_6B-fim"5device ="cuda"# for GPU usage or "cpu" for CPU usage67tokenizer = AutoTokenizer.from_pretrained(checkpoint)8model = AutoModelForCausalLM.from_pretrained(checkpoint, trust_remote_code=True).to(device)910prompt ='<fim_prefix>def print_hello_world():\n """<fim_suffix>\n print("Hello world!")<fim_middle>'1112inputs = tokenizer.encode(prompt, return_tensors="pt").to(device)13outputs = model.generate(inputs, max_length=100, temperature=0.2)14print("-"*80)15print(tokenizer.decode(outputs[0]))
Chat Format
The same model works as chat (experimental).
python
1prompt_template ="<empty_output>SYSTEM {system}\n" \
2"<empty_output>USER {query}\n" \
3"<empty_output>ASSISTANT"4prompt = prompt_template.format(system="You are a programming assistant",5 query="How do I sort a list in Python?")
Architecture
As described in more detail in the blog post, we used:
We also used LiON, flash attention, early dropout. It's not that innovative that you can't run it, in fact you can -- see an example below.
Training
For the base model, we used our own dataset that contains code with permissive licenses only, and open text datasets.
Filtering is the key to success of this model:
We only used text in English
Only topics related to computer science
Applied heavy deduplication
The text to code proportion was 50:50, model trained for 1.2T tokens.
We don't release the base model, because its Fill-in-the-Middle (FIM) capability likes to repeat itself too much, so
its practical use is limited. But if you still want it, write us a message on Discord.
Limitations and Bias
The Refact-1.6B model was trained on text in English. But it has seen a lot more languages in
code comments. Its performance on non-English languages is lower, for sure.
Model Stats
Architecture: LLAMA-like model with multi-query attention
Objectives Fill-in-the-Middle, Chat
Tokens context: 4096
Pretraining tokens: 1.2T
Finetuning tokens: 40B
Precision: bfloat16
GPUs 64 NVidia A5000
Training time 28 days
License
The model is licensed under the BigScience OpenRAIL-M v1 license agreement
Citation
If you are using this model, please give a link to this page.
