trurl-2-13b-8bit

<img src="https://public.3.basecamp.com/p/rs5XqmAuF1iEuW6U7nMHcZeY/upload/download/VL-NLP-short.png" alt="logo voicelab nlp" style="width:300px;"/>

Trurl 2 -- Polish Llama 2

The new OPEN TRURL is a finetuned Llama 2, trained on over 1.7b tokens (970k conversational Polish and English samples) with a large context of 4096 tokens. TRURL was trained on a large number of Polish data. TRURL 2 is a collection of fine-tuned generative text models with 7 billion and 13 billion parameters.

ATTENTION

This is the repository for the 13b fine-tuned QUANTIZED model 8-bit, optimized for dialogue use cases. Quantization results in a significantly smaller and faster model at a cost of slightly (or sometimes considerably) worse results, depending on the size reduction and provided input.

This models takes around 14Gb of GPU RAM.

If you need a full version check it here:

• TRURL 13b
• TRURL 7b

Overview

TRURL developers Voicelab.AI

Variations Trurl 2 comes in 7B and 13B versions.

Input Models input text only.

Output Models generate text only.

Model Architecture Trurl is an auto-regressive language model that uses an optimized transformer architecture.

Training data

The training data includes Q&A pairs from various sources including Alpaca comparison data with GPT, Falcon comparison data, Dolly 15k, Oasst1, Phu saferlfhf, ShareGPT version 2023.05.08v0 filtered and cleaned, Voicelab private datasets for JSON data extraction, modification, and analysis, CURLICAT dataset containing journal entries, dataset from Polish wiki with Q&A pairs grouped into conversations, MMLU data in textual format, Voicelab private dataset with sales conversations, arguments and objections, paraphrases, contact reason detection, and corrected dialogues.

Intended Use

Trurl 2 is intended for commercial and research use in Polish and English. Tuned models are intended for assistant-like chat, but also adapted for a variety of natural language generation tasks.

Evaluation Results

<img src="https://voicelab.ai/wp-content/uploads/trurl-hero.webp" alt="trurl graphic" style="width:100px;"/>

Ethical Considerations and Limitations

Trurl 2, same as a Llama 2, is a new technology that carries risks with use. Testing conducted to date has been in Polish and English, and has not covered, nor could it cover all scenarios. For these reasons, as with all LLMs, Trurl 2’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of Trurl 2, developers should perform safety testing and tuning tailored to their specific applications of the model.

Please see the Meta's Responsible Use Guide available at https://ai.meta.com/llama/responsible-use-guide/

Example use

Installation

To use Quantized models you have to have the newest transformers (pip install transformers --upgrade), tokenizers (pip install tokenizers --upgrade), accelerate (pip install accelerate --upgrade) and bitsandbytes (pip install accelerate --bitsandbytes).

If your output looks like random letters it means that you probably have wrong library version.

LLM

Simply pass a prompt to a model and decode an output. Model will continue writing text based on sample you provided.

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("Voicelab/trurl-2-13b-8bit")
model = AutoModelForCausalLM.from_pretrained("Voicelab/trurl-2-13b-8bit", device_map="auto")

prompt = "Yesterday, when I was"

tokenized_prompt = tokenizer(prompt, return_tensors="pt")

model.eval()
with torch.no_grad():
    print(tokenizer.decode(
        model.generate(tokenized_prompt.data["input_ids"], max_new_tokens=200, temperature=0)[0],
        skip_special_tokens=True))

Generated output:

Yesterday, when I was a guest on the "Today" show, I was asked about the possibility of a government shutdown. I said that I believed it was unlikely, but that it was up to the politicians to decide. I was surprised by the reaction of some viewers, who took my statement as a sign that I was in favor of a shutdown. Nothing could be further from the truth. I am strongly opposed to a government shutdown, and I believe that it would be a disaster for our country. As I have said before, the decision to shut down the government should not be taken lightly. It would have serious consequences for our economy and for the millions of Americans who rely on government services. It would also be a breach of trust with the American people, who expect their elected officials to work together to address the challenges facing our nation. I will continue to urge my colleagues in Congress to find a way to fund the government

Chat

When using TRURL in a chat mode you should remember to use Llama 2 conversation template like in the example below.

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("Voicelab/trurl-2-13b-8bit")
model = AutoModelForCausalLM.from_pretrained("Voicelab/trurl-2-13b-8bit", device_map="auto")

prompt = """
<s>[INST] <<SYS>>  You are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe.
Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content.
Please ensure that your responses are socially unbiased and positive in nature.\n\n
If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct.
If you don't know the answer to a question, please don't share false information. <</SYS>>

What was the reason for calling in the conversation below? \n\n
AGENT: Hello, Bank of Albion, this is Mata Hari. How can I help you?
CLIENT: Hi. I've been locked out from my Internet account. I need your help.
AGENT: (yy) Yes, of course, I'll do my best to help you. But I need to find out why the locking-out happened. (yy) In order to ascertain that, I'll ask you a couple of questions to confirm your identity. I'm going to need your full name.
CLIENT: Lizz Truss.
AGENT: Thank you. Now I need your personal identification number.
CLIENT: Fourteen, two hundred thirty-one, thirty-eight, twenty-nine, sixty-five.
AGENT: Thank you. Now I need your client ID number. The client ID number is the eight digits we assigned to you at the very beginning, on conclusion of the contract.
CLIENT: OK. Give me a moment. I have to find it.
AGENT: (mhm) You'll find… You'll find it in the contract.
CLIENT: Yes, yes. I can see it. Sixty-five, twenty-nine, thirty-eight, thirty-one.
AGENT: Thank you. One final security question. Do you have any deposits in our bank?
CLIENT: No, no. I don't have any deposits in this bank.
AGENT: Thank you. Your identity has been (yy) confirmed. (yy) I can see that the account has been blocked, indeed, and you won't be able to log in via the Internet (yy) because (yy) the identity document which is listed for reference has expired. (yy) From what I can see, your identity document expired some time ago. Have you been issued a new one?
CLIENT: Well, no. I think my ID is still valid, you know. I didn't even know.
AGENT: Well, no... Your ID expired at the end of March. Well, almost at the end. Your old ID had been valid until 26 March. (yy) For that reason, your accout has been blocked, because you haven't notified us about the ID change for a few months. We are not interested if the ID document has been officialy reissued. (...) On our end, what matters is whether the document listed for our reference is valid (yy) so without a valid document I can't unlock your accout. 
CLIENT: But I have to carry out an operation right now, so this is sort of problematic.
AGENT: I understand. But (yy) you are obligated, as an account holder, to notify the bank about any changes pending (yy), regrding, for example, your home address or phone number. Now, one of such safeguards protecting your… (yy) money, your sensitive data, is precisely about having a valid identification document. Since this is missing in your case, the account has been blocked. Now, I don't think this would have caught you off guard, because we always remind our customers that their ID is about to expire. When the ID is nearing expiration, we display relevant messages at least sixty days in advance. They appear once you've logged in, at the very top of the screen, there is a notification that (yy) the ID is about to expire (yy), so, well... The bank did notify you about this issue. Now, how you chose to act on this information was your choice, right? In any case, at this point, in order to unlock your accout, our protocols require that you produce a new identification document at one of our branches. You shall provide information concerning the new document number, new valid-thru date, and only then will you be able to use your account again. I can schedule an appointment with a consultant at our branch for you. What locality would you prefer?
CLIENT: Well, I'm not sure if I should share such information with you.
AGENT: And may I ask why exactly you are unsure? After all, you're calling a bank that runs your account, right?
CLIENT: Right, you know what, I need to go now. Good bye.
AGENT: (yy) Miss… [/INST]

"""

tokenized_prompt = tokenizer(prompt, return_tensors="pt")

model.eval()
with torch.no_grad():
    print(tokenizer.decode(
        model.generate(tokenized_prompt.data["input_ids"], max_new_tokens=200, temperature=0)[0],
        skip_special_tokens=True))

Generated output:

The reason for calling was for the client to request help with accessing their Internet account, which had been locked out. The agent asked questions to confirm the client's identity and discovered that the account had been blocked because the client's identification document had expired. The agent explained that the bank had notified the client about the issue and that in order to unlock the account, the client needed to provide a new identification document at one of the bank's branches. The agent offered to schedule an appointment for the client and asked for their preferred location. The client then ended the call without providing any further information.

To get the expected features and performance for the chat versions, a specific Llama 2 formatting needs to be followed, including the INST and <<SYS>> tags, BOS and EOS tokens, and the whitespaces and breaklines in between (we recommend calling strip() on inputs to avoid double-spaces). See reference code in github for details: chat_completion.

Authors

The model was trained by NLP Research Team at Voicelab.ai.

You can contact us here.

• TRURL 13b
• TRURL 13b Academic
• TRURL 7b
• TRURL DEMO

Quantized models:

• TRURL 13b - 8bit
• TRURL 7b - 8bit

The work was supported by #NASK