metadata
library_name: transformers
license: apache-2.0
datasets:
- gretelai/synthetic_text_to_sql
- NickyNicky/synthetic_text_to_sql_format_chatML_gemma
language:
- en
widget:
- text: >
<bos><start_of_turn>system
You are a helpful AI assistant.
you are a sql expert who responds in json format.<end_of_turn>
<start_of_turn>user
## prompt:
What is the total gold production by 'Site B' in the 'production' table?
## sql context:
CREATE TABLE production (id INT, site VARCHAR(50), year INT,
gold_production INT, silver_production INT);<end_of_turn>
<start_of_turn>model
Metrics.
TrainOutput(global_step=2509,
training_loss=0.2509715025906736,
metrics={'train_runtime': 22783.0743,
'train_samples_per_second': 8.81,
'train_steps_per_second': 0.11,
'total_flos': 1.820581902144553e+18,
'train_loss': 0.2509715025906736,
'epoch': 2.01
}
)
Take dataset.
gretelai/synthetic_text_to_sql
Dataset format gemma fine tune.
NickyNicky/synthetic_text_to_sql_format_chatML_gemma
colab examples and Gradio.
https://colab.research.google.com/drive/1-0PsRAqTum2UuvsXb9JtXxLISIUWd8zv?usp=sharing
gradio colab.
$ train