After 150+ hours of training, HTGM v2 Hindi LLM was tested using real Hindi question-answer prompts to evaluate how the base model is learning.
This is still the base model stage — not fine-tuned yet — and only around 4% of total planned training has been completed.
The goal of this test was simple:
To check whether the model is learning factual patterns, language structure, and question-answer understanding.
This repository documents the real outputs, observations, and current model status.
- Model Name: HTGM v2
- Type: Hindi Large Language Model (Hindi LLM)
- Architecture: GPT-style Transformer
- Parameters: 163.4M+
- Sequence Length: 2048
- Vocabulary Size: 100,000
- Dataset Size: ~41GB
- Dataset Source: AI4Bharat Sangrah + Manual Cleaning
- Training Platform: Kaggle
- GPU: 2x T4
- Tokenizer: Hugging Face BPE-based Tokenizer
- Training Started: 1 April
- Total Training Time: 150+ Hours
- Total Days: 24 Days
- Sessions Completed: 17
- Optimizer Steps: 38K+
- Total Training Completion: ~4%
- Training Still Running
This means the model is still in an early-stage pretraining phase.
A lot more learning is still left.
- Model Loader: GPT2LMHeadModel
- Device: CPU
- Tokenizer: Custom BPE Tokenizer
- temperature: 0.7
- top_k: 50
- top_p: 0.9
- repetition_penalty: 1.2
These settings were used to reduce repetition and improve generation stability.
These are real outputs directly generated by HTGM v2.
No manual editing.
प्रश्न: हिंदी भाषा किस देश में बोली जाती है? उत्तर:
The model understood the topic was related to language and words, but failed to directly answer “भारत”.
It generated broken structure and drifted into unrelated patterns.
Pattern recognition exists, but factual answering is still weak.
प्रश्न: सूर्य किस दिशा से निकलता है? उत्तर:
The model understood the concept of direction and time, but failed to correctly answer “पूर्व दिशा”.
Instead, it generated repetitive reasoning.
Context recognition exists, but answer completion is unreliable.
प्रश्न: भारत की राजधानी क्या है? उत्तर:
The model correctly started with:
दिल्ली
…but then drifted into unrelated repetitive text.
This is a strong sign that factual learning has started.
This is an important milestone.
प्रश्न: तुम कैसे हो? उत्तर:
The model understood conversational format but entered heavy repetition like:
“बस बस बस…”
Conversation structure is learned, but coherence is still weak.
This is not failure.
This is exactly how base model learning works.
At this stage, success means:
- Correct starting tokens
- Pattern recognition
- Q&A format understanding
- Partial factual recall
HTGM v2 is already showing these signs.
That is real progress.
- High repetition
- Weak coherence
- Context drifting
- Incomplete factual answers
- No instruction fine-tuning yet
This is expected because the model is still only around 4% trained.
- More pretraining
- Supervised Fine-Tuning (SFT)
- Better dataset structuring
- Stronger instruction-following
- Reduced repetition
- Better output alignment
This is where major quality improvements will happen.
After 150+ hours, HTGM v2 is not perfect.
But it is learning.
It recognizes patterns.
It starts correct answers.
It understands structure.
It shows real progress.
This is still the beginning — not the final model.
But the foundation is becoming real.
Mahesh Editor
India AI Official
No fake hype.
Only real outputs.
Only real progress.
This is how real AI gets built.