Text Generation
Transformers
Safetensors
15 languages
gemma
Inference Endpoints
text-generation-inference
Edit model card

Indic-gemma-7b-finetuned-sft-Navarasa-2.0

This model is based on google/gemma-7b and hase been LoRA finetuned on 15 Indian languages and English language instruction datasets:

  1. Hindi - ravithejads/samvaad-hi-filtered, HydraIndicLM/hindi_alpaca_dolly_67k(sampled)

  2. Telugu - Telugu-LLM-Labs/telugu_alpaca_yahma_cleaned_filtered_romanized, Telugu-LLM-Labs/telugu_teknium_GPTeacher_general_instruct_filtered_romanized

  3. Marathi - Telugu-LLM-Labs/sindhi_alpaca_yahma_cleaned_filtered

  4. Urdu - Telugu-LLM-Labs/urdu_alpaca_yahma_cleaned_filtered

  5. Assamese - Telugu-LLM-Labs/assamese_alpaca_yahma_cleaned_filtered

  6. Konkani - Telugu-LLM-Labs/konkani_alpaca_yahma_cleaned_filtered

  7. Nepali - Telugu-LLM-Labs/nepali_alpaca_yahma_cleaned_filtered

  8. Sindhi - Telugu-LLM-Labs/sindhi_alpaca_yahma_cleaned_filtered

  9. Tamil - abhinand/tamil-alpaca

  10. Kannada - Tensoic/airoboros-3.2_kn, Tensoic/gpt-teacher_kn

  11. Malayalam - VishnuPJ/Alpaca_Instruct_Malayalam

  12. Gujarati - Tensoic/Alpaca-Gujarati

  13. Punjabi - HydraIndicLM/punjabi_alpaca_52K

  14. Bengali - HydraIndicLM/bengali_alpaca_dolly_67k(alpaca filtered)

  15. Odia - OdiaGenAI/Odia_Alpaca_instructions_52k, OdiaGenAI/gpt-teacher-roleplay-odia-3k

  16. English - yahma/alpaca-cleaned

The model is finetuned using unsloth library and we provide inference code using the same for faster inference. Alternatively you can use HuggingFace Library for inference.

Training Details:

The model is trained on approx 650K instruction samples.

  1. GPU: 1 A100, 80GB
  2. Time: 45 Hours
  3. Platform: E2E Networks

Installation

!pip install -U xformers --index-url https://download.pytorch.org/whl/cu121 !pip install "unsloth[kaggle-new] @git+https://github.com/unslothai/unsloth.git@nightly"

Input Text Format

### Instruction: {instruction}

### Input: {input}

## Response: {response}

Inference With Unsloth

from unsloth import FastLanguageModel
import torch
max_seq_length = 2048
dtype = None # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+
load_in_4bit = False 
model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "Telugu-LLM-Labs/Indic-gemma-7b-finetuned-sft-Navarasa-2.0",
    max_seq_length = max_seq_length,
    dtype = dtype,
    load_in_4bit = load_in_4bit,
    device_map="auto"
)
FastLanguageModel.for_inference(model) # Enable native 2x faster inference

input_prompt = """
### Instruction:
{}

### Input:
{}

### Response:
{}"""

input_text = input_prompt.format(
        "Tranlsate following sentence to Hindi.", # instruction
        "India is a great country.", # input
        "", # output - leave this blank for generation!
    )

inputs = tokenizer([input_text], return_tensors = "pt").to("cuda")

outputs = model.generate(**inputs, max_new_tokens = 300, use_cache = True)
response = tokenizer.batch_decode(outputs)

Inference with HuggingFace

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model = AutoModelForCausalLM.from_pretrained(
    "Telugu-LLM-Labs/Indic-gemma-7b-finetuned-sft-Navarasa-2.0",
    load_in_4bit = False,
    token = hf_token
)
model.to("cuda")

tokenizer = AutoTokenizer.from_pretrained("Telugu-LLM-Labs/Indic-gemma-7b-finetuned-sft-Navarasa-2.0")

input_prompt = """
### Instruction:
{}

### Input:
{}

### Response:
{}"""

input_text = input_prompt.format(
        "Tranlsate following sentence to Hindi.", # instruction
        "India is a great country.", # input
        "", # output - leave this blank for generation!
    )

inputs = tokenizer([input_text], return_tensors = "pt").to("cuda")

outputs = model.generate(**inputs, max_new_tokens = 300, use_cache = True)
response = tokenizer.batch_decode(outputs)[0]

Refer to the blog post for sample examples.

Please check our Code Repository for training and inference scripts.

Developers:

The model is a collaborative effort by Ravi Theja and Ramsri Goutham. Feel free to DM either of us if you have any questions.

Downloads last month
1,145
Safetensors
Model size
8.54B params
Tensor type
F32
·

Finetuned from

Datasets used to train Telugu-LLM-Labs/Indic-gemma-7b-finetuned-sft-Navarasa-2.0

Spaces using Telugu-LLM-Labs/Indic-gemma-7b-finetuned-sft-Navarasa-2.0 2

Collection including Telugu-LLM-Labs/Indic-gemma-7b-finetuned-sft-Navarasa-2.0