Indic Gemma 2B Finetuned Sft Navarasa 2.0 By Telugu-LLM-Labs: Benchmarks, Features and Detailed Analysis. Insights on Indic Gemma 2B Finetuned Sft Navarasa 2.0.

As Autotrain compatible Base model:finetune:google/gem... Base model:google/gemma-2b Bn Dataset:abhinand/tamil-alpaca Dataset:hydraindiclm/bengali a... Dataset:hydraindiclm/punjabi a... Dataset:odiagenai/odia alpaca ... Dataset:ravithejads/samvaad-hi... Dataset:telugu-llm-labs/assame... Dataset:telugu-llm-labs/konkan... Dataset:telugu-llm-labs/marath... Dataset:telugu-llm-labs/nepali... Dataset:telugu-llm-labs/sindhi... Dataset:telugu-llm-labs/telugu... Dataset:telugu-llm-labs/telugu... Dataset:telugu-llm-labs/urdu a... Dataset:tensoic/airoboros-3.2 ... Dataset:tensoic/alpaca-gujarat... Dataset:tensoic/gpt-teacher kn Dataset:vishnupj/alpaca instru... Dataset:yahma/alpaca-cleaned En Endpoints compatible Finetuned Gemma Gu Hi Instruct Kn Ml Mr Ne Or Pa Region:us Safetensors Sd Sharded Ta Te Tensorflow Ur

Model Card on HF 🤗: https://huggingface.co/Telugu-LLM-Labs/Indic-gemma-2b-finetuned-sft-Navarasa-2.0

Indic Gemma 2B Finetuned Sft Navarasa 2.0 Benchmarks

ARC: 44.71 vs 96.7 (so35)^-53.8%

HellaSwag: 68.4 vs 95.3 (gpt4)^-28.2%

MMLU: 38.21 vs 88.3 (so35)^-56.7%

TruthfulQA: 44.69 vs 59 (gpt4)^-24.3%

WinoGrande: 65.11 vs 87.5 (gpt4)^-25.6%

GSM8K: 9.25 vs 96.4 (so35)^-90.4%

LLME Score: 0.23838

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

Indic Gemma 2B Finetuned Sft Navarasa 2.0 (Telugu-LLM-Labs/Indic-gemma-2b-finetuned-sft-Navarasa-2.0)

Indic Gemma 2B Finetuned Sft Navarasa 2.0 Parameters and Internals

Model Type

text generation

Additional Notes

The model can also be inferred using the HuggingFace Library or Unsloth library for faster inference.

Supported Languages

te (Telugu), en (English), ta (Tamil), ml (Malayalam), mr (Marathi), hi (Hindi), kn (Kannada), sd (Sindhi), ne (Nepali), ur (Urdu), as (Assamese), gu (Gujarati), bn (Bengali), pa (Punjabi), or (Odia)

Training Details

Data Sources:

ravithejads/samvaad-hi-filtered, Telugu-LLM-Labs/telugu_teknium_GPTeacher_general_instruct_filtered_romanized, Telugu-LLM-Labs/telugu_alpaca_yahma_cleaned_filtered_romanized, Telugu-LLM-Labs/sindhi_alpaca_yahma_cleaned_filtered, Telugu-LLM-Labs/urdu_alpaca_yahma_cleaned_filtered, Telugu-LLM-Labs/marathi_alpaca_yahma_cleaned_filtered, Telugu-LLM-Labs/assamese_alpaca_yahma_cleaned_filtered, Telugu-LLM-Labs/konkani_alpaca_yahma_cleaned_filtered, Telugu-LLM-Labs/nepali_alpaca_yahma_cleaned_filtered, abhinand/tamil-alpaca, Tensoic/airoboros-3.2_kn, Tensoic/gpt-teacher_kn, VishnuPJ/Alpaca_Instruct_Malayalam, Tensoic/Alpaca-Gujarati, HydraIndicLM/punjabi_alpaca_52K, HydraIndicLM/bengali_alpaca_dolly_67k, OdiaGenAI/Odia_Alpaca_instructions_52k, yahma/alpaca-cleaned

Data Volume:

approx 650K instruction samples

Methodology:

LoRA finetuning on instruction datasets

Context Length:

2048

Training Time:

45 Hours

Hardware Used:

1 A100, 80GB GPU

Input Output

Input Format:

### Instruction: {instruction} ### Input: {input} ## Response: {response}

LLM Name	Indic Gemma 2B Finetuned Sft Navarasa 2.0
Repository 🤗	https://huggingface.co/Telugu-LLM-Labs/Indic-gemma-2b-finetuned-sft-Navarasa-2.0
Base Model(s)	Gemma 2B google/gemma-2b
Model Size	2b
Required VRAM	10 GB
Updated	2024-12-22
Maintainer	Telugu-LLM-Labs
Model Type	gemma
Instruction-Based	Yes
Model Files	4.9 GB: 1-of-3 5.0 GB: 2-of-3 0.1 GB: 3-of-3
Supported Languages	te en ta ml mr hi kn sd ne ur as gu bn pa or
Model Architecture	GemmaForCausalLM
License	other
Context Length	8192
Model Max Length	8192
Transformers Version	4.38.2
Tokenizer Class	GemmaTokenizer
Padding Token	<pad>
Vocabulary Size	256000
Torch Data Type	float32

Quantized Models of the Indic Gemma 2B Finetuned Sft Navarasa 2.0

Model	Likes	Downloads	VRAM
...B Malayalam T2 Model Vllm 4bit	0	18	2 GB
Gemma 2B Malayalam T2 Gguf	0	17	1 GB

Best Alternatives to Indic Gemma 2B Finetuned Sft Navarasa 2.0

Best Alternatives	Context / RAM	Downloads	Likes
Leniachat Gemma 2B V0	8K / 15.1 GB	560	12
...ma Instruct Finetune On Alpaca	8K / 5.1 GB	10	0
...ma Code Instruct Finetune Test	8K / 5.1 GB	10	0
Gemma 2B It Customer Support	8K / 10 GB	495	1
Gemma 2B Persona Instruct	8K / 5.1 GB	17	0
...ma Code Instruct Finetune Test	8K / 5.1 GB	14	0
Gemma 2B Sft Preview	8K / 5 GB	22	0
...ma Code Instruct Finetune Test	8K / 5.1 GB	10	0
Datagemma 2B	8K / 5.1 GB	17	0
...truct Finetune Test Mf Tuned 2	8K / 5.1 GB	19	2

Rank the Indic Gemma 2B Finetuned Sft Navarasa 2.0 Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

What open-source LLMs or SLMs are you in search of? 40123 in total.

Email us: info@extractum.io. Our Privacy Policy | Terms and Conditions | Suggest an improvement.

Our Social Media →

Original data from HuggingFace, OpenCompass and various public git repos.

Release v20241217

Support LLM Explorer