VictorDCh's picture
Update README.md
ad12a50 verified
|
raw
history blame
1.91 kB
metadata
license: llama3
library_name: peft
tags:
  - trl
  - sft
  - generated_from_trainer
base_model: meta-llama/Meta-Llama-3-8B-Instruct
datasets:
  - generator
model-index:
  - name: Llama-3-8B-Instruct-2-spider-3
    results: []

Llama-3-8B-Instruct-2-spider-3

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on the generator dataset.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

prompt format: """You are a helpful assistant who answers questions about database tables by responding with SQL queries. Users will provide you with a set of tables represented as CREATE TABLE statements followed by INSERT statements containing information about the data types contained in each of the tables. Then users will ask a question. Answer the user question by writing a SQL statement with the help of the provided SQL tables."""

user_message = """Here is the schema of the tables you will be working with:

{schema}

-- {question} SELECT"""

dataset: VictorDCh/spider-clean-text-to-sql-3

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 0.0002
  • train_batch_size: 1
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 2
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.03
  • num_epochs: 1

Training results

Framework versions

  • PEFT 0.7.2.dev0
  • Transformers 4.36.2
  • Pytorch 2.1.2+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.2