Instructions to use GarvinBhati/gemma-4-31b-indic-rys with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use GarvinBhati/gemma-4-31b-indic-rys with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("google/gemma-4-31B-it") model = PeftModel.from_pretrained(base_model, "GarvinBhati/gemma-4-31b-indic-rys") - Notebooks
- Google Colab
- Kaggle
Download evaluation/generation_audit.csv from GarvinBhati/gemma-4-31b-indic-rys: direct link, hf CLI and curl.
- Browser
- Download file 741 Bytes
-
https://huggingface.co/GarvinBhati/gemma-4-31b-indic-rys/resolve/main/evaluation/generation_audit.csv
- Command line
-
hf download hf://GarvinBhati/gemma-4-31b-indic-rys/evaluation/generation_audit.csv
-
curl -L -o generation_audit.csv https://huggingface.co/GarvinBhati/gemma-4-31b-indic-rys/resolve/main/evaluation/generation_audit.csv
741 Bytes
| evaluation_key,benchmark,rows,output_cap,output_tokens,mean_output_tokens,length_capped,length_capped_rate | |
| stock_base,milu,3980,8,8638,2.17035175879397,113,0.028391959798994976 | |
| stock_base,indicparam,660,8,1332,2.018181818181818,2,0.0030303030303030303 | |
| stock_base,indicifeval,602,2048,232636,386.4385382059801,2,0.0033222591362126247 | |
| selected_sft,milu,3980,8,8062,2.0256281407035175,17,0.004271356783919598 | |
| selected_sft,indicparam,660,8,1320,2.0,0,0.0 | |
| selected_sft,indicifeval,602,4096,322238,535.2790697674419,33,0.054817275747508304 | |
| sft_plus_gated,milu,3980,8,8044,2.021105527638191,14,0.0035175879396984926 | |
| sft_plus_gated,indicparam,660,8,1320,2.0,0,0.0 | |
| sft_plus_gated,indicifeval,602,4096,316750,526.1627906976744,32,0.053156146179401995 | |