How to fine-tune an LLM with AutoTrain?

🤗AutoTrain
1

Hi,
This document indicates that one can fine-tune an LLM with AutoTrain using CLM.
I have a dataset that is formatted as:
{ “instruction”: “xxx”, “input”: “yyy”, “output”: “zzz”} tuples.

When attempting to create a new AutoTrain project, I’m not sure which option to choose to be able to train the model with these tuples using CLM.

Any suggestions would be helpful. Thank you!

2

you seem to have a jsonl file, you can convert it to CSV using python:

import pandas as pd

# read the JSONL file into a pandas DataFrame
df = pd.read_json('input.jsonl', lines=True)

# write the DataFrame to a CSV file
df.to_csv('output.csv', index=False)

regarding column mapping, it seems almost the same example is provided as an example:

image
image2994Ă—1562 310 KB

3

Sorry, I should have pasted this picture.
Which of the model types map to training the model with CLM?
image

4

llm is only available in autotrain advanced: AutoTrain

5

Are there specific resource requirements for the hugging face space environment in order to proceed with LLM Finetuning?

1 Like
6

Hi @abhishek ,
I have a follow-up question please.

According to the Hugging Face documentation, this is the format for SFT: human: ... bot: ...

image
image1315Ă—820 57.6 KB

However, according to the example dataset the string in the text column should be in the following format:
### Human: ... ### Assistant: ...

When opening a new AutoTrain project, the following column mapping is requested: {“text”: “text”}

So, my questions:

  1. Should it be ### Human: ... ### Assistant: ... or human: ... bot: ...? (human or Human, Assistant or bot, case-senstive, with/out ###?)
  2. I also so in the example dataset a chaining of human-assistant pairs. In what case would I use this instead of just one pair?
  3. What do I do if I want to add context? Basically, I want to fine-tune a model in order to create a custom chatbot that is fine-tuned on conversations. Therefore, I need to input not only a single question and and answer but rather add all the conversation that preceded a specific question as context.
  4. Is there a way to input a chain-of-thought input in the form of a JSON? For example: {"conversation_context": conversation_context, "customer_question": customer_question, "assistant_response": assistant_response}?

Many thanks :pray:t3: