Modelpublic

RichardErkhov/NickyNicky_-_gemma-1.1-2b-it_text_to_sql_format_chatML_V1-gguf

sourceHugging Faceupdated 2y agoView on Hugging Face

0likes552downloads

Model Card

Quantization made by Richard Erkhov.

Github

Discord

Request more models

gemma-1.1-2b-ittexttosqlformatchatMLV1 - GGUF

—Model creator: https://huggingface.co/NickyNicky/
—Original model: https://huggingface.co/NickyNicky/gemma-1.1-2b-ittexttosqlformatchatMLV1/

Name	Quant method	Size
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q2_K.gguf	Q2_K	1.08GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.IQ3_XS.gguf	IQ3_XS	1.16GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.IQ3_S.gguf	IQ3_S	1.2GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q3_K_S.gguf	Q3KS	1.2GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.IQ3_M.gguf	IQ3_M	1.22GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q3_K.gguf	Q3_K	1.29GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q3_K_M.gguf	Q3KM	1.29GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q3_K_L.gguf	Q3KL	1.36GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.IQ4_XS.gguf	IQ4_XS	1.4GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q4_0.gguf	Q4_0	1.44GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.IQ4_NL.gguf	IQ4_NL	1.45GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q4_K_S.gguf	Q4KS	1.45GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q4_K.gguf	Q4_K	1.52GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q4_K_M.gguf	Q4KM	1.52GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q4_1.gguf	Q4_1	1.56GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q5_0.gguf	Q5_0	1.68GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q5_K_S.gguf	Q5KS	1.68GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q5_K.gguf	Q5_K	1.71GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q5_K_M.gguf	Q5KM	1.71GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q5_1.gguf	Q5_1	1.79GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q6_K.gguf	Q6_K	1.92GB
gemma-1.1-2b-it_text_to_sql_format_chatML_V1.Q8_0.gguf	Q8_0	2.49GB

Original model description: --- library_name: transformers license: apache-2.0 datasets:

—gretelai/synthetictextto_sql
—NickyNicky/synthetictexttosqlformatchatMLgemma language:
—en

widget:

—text: | <bos><startofturn>system You are a helpful AI assistant. you are a sql expert who responds in json format.<endofturn> <startofturn>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, goldproduction INT, silverproduction INT);<endofturn> <startofturn>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