---
title: "#ai-yatra articles | Hashnode"
description: "Explore 77 articles tagged #ai-yatra on Hashnode. Tutorials, guides, and how-tos from the community."
url: "https://hashnode.com/tag/ai-yatra"
---

Tag

# #ai-yatra

77 posts · 0 followers

[Pankaj Kumar](https://hashnode.com/@pankajhyd) · [Aug 22, 2025](https://hashnode.com/posts/vision-yatra-step-10-layer-normalization-and-residual-connections-stabilizing-the-transformer/68a8595e44e58f2fe94e1600)

### [Vision Yatra: Step 10 — Layer Normalization & Residual Connections: Stabilizing the Transformer](https://my-ai-yatra.solveautomation.in/vision-yatra-step-10-layer-normalization-and-residual-connections-stabilizing-the-transformer?utm_source=hashnode&utm_medium=feed)

How Transformers avoid vanishing gradients and train deep networks

5 min read · [my-ai-yatra.solveautomation.in](https://my-ai-yatra.solveautomation.in/vision-yatra-step-10-layer-normalization-and-residual-connections-stabilizing-the-transformer?utm_source=hashnode&utm_medium=feed)

[Pankaj Kumar](https://hashnode.com/@pankajhyd) · [Aug 22, 2025](https://hashnode.com/posts/vision-yatra-step-16-the-final-layer-linear-and-softmax-how-transformers-generate-words/68a86cbb4474a7e82250b21a)

### [Vision Yatra: Step 16 — The Final Layer: Linear & Softmax — How Transformers Generate Words](https://my-ai-yatra.solveautomation.in/vision-yatra-step-16-the-final-layer-linear-and-softmax-how-transformers-generate-words?utm_source=hashnode&utm_medium=feed)

From vectors to words — the secret behind text generation in ChatGPT and BERT

6 min read · [my-ai-yatra.solveautomation.in](https://my-ai-yatra.solveautomation.in/vision-yatra-step-16-the-final-layer-linear-and-softmax-how-transformers-generate-words?utm_source=hashnode&utm_medium=feed)

[Pankaj Kumar](https://hashnode.com/@pankajhyd) · [Aug 22, 2025](https://hashnode.com/posts/vision-yatra-step-13-the-transformer-decoder-how-models-generate-text-one-token-at-a-time/68a8634ffd65fd91f63a05cd)

### [Vision Yatra: Step 13 — The Transformer Decoder: How Models Generate Text One Token at a Time](https://my-ai-yatra.solveautomation.in/vision-yatra-step-13-the-transformer-decoder-how-models-generate-text-one-token-at-a-time?utm_source=hashnode&utm_medium=feed)

From masked attention to text generation — the secret behind ChatGPT's output

5 min read · [my-ai-yatra.solveautomation.in](https://my-ai-yatra.solveautomation.in/vision-yatra-step-13-the-transformer-decoder-how-models-generate-text-one-token-at-a-time?utm_source=hashnode&utm_medium=feed)

[Pankaj Kumar](https://hashnode.com/@pankajhyd) · [Aug 22, 2025](https://hashnode.com/posts/vision-yatra-step-12-the-transformer-encoder-full-architecture-breakdown/68a85ffcc1336db3bfd622af)

### [Vision Yatra: Step 12 — The Transformer Encoder: Full Architecture Breakdown](https://my-ai-yatra.solveautomation.in/vision-yatra-step-12-the-transformer-encoder-full-architecture-breakdown?utm_source=hashnode&utm_medium=feed)

From 512D vectors to 6 stacked layers — how the encoder processes language

6 min read · [my-ai-yatra.solveautomation.in](https://my-ai-yatra.solveautomation.in/vision-yatra-step-12-the-transformer-encoder-full-architecture-breakdown?utm_source=hashnode&utm_medium=feed)

[1](https://hashnode.com/posts/vision-yatra-step-12-the-transformer-encoder-full-architecture-breakdown/68a85ffcc1336db3bfd622af)

[Pankaj Kumar](https://hashnode.com/@pankajhyd) · [Aug 22, 2025](https://hashnode.com/posts/vision-yatra-step-15-encoder-decoder-attention-how-transformers-translate-between-languages/68a86a1328817b4e90debfab)

### [Vision Yatra: Step 15 — Encoder-Decoder Attention: How Transformers "Translate" Between Languages](https://my-ai-yatra.solveautomation.in/vision-yatra-step-15-encoder-decoder-attention-how-transformers-translate-between-languages?utm_source=hashnode&utm_medium=feed)

The secret behind alignment — how models know which input words to focus on for each output word

6 min read · [my-ai-yatra.solveautomation.in](https://my-ai-yatra.solveautomation.in/vision-yatra-step-15-encoder-decoder-attention-how-transformers-translate-between-languages?utm_source=hashnode&utm_medium=feed)

[Pankaj Kumar](https://hashnode.com/@pankajhyd) · [Aug 22, 2025](https://hashnode.com/posts/vision-yatra-step-14-masked-multi-head-self-attention-how-transformers-generate-text-sequentially/68a8662f7942a23a6043dd7d)

### [Vision Yatra: Step 14 — Masked Multi-Head Self-Attention: How Transformers Generate Text Sequentially](https://my-ai-yatra.solveautomation.in/vision-yatra-step-14-masked-multi-head-self-attention-how-transformers-generate-text-sequentially?utm_source=hashnode&utm_medium=feed)

The secret behind ChatGPT's word-by-word generation — no peeking at the future!

5 min read · [my-ai-yatra.solveautomation.in](https://my-ai-yatra.solveautomation.in/vision-yatra-step-14-masked-multi-head-self-attention-how-transformers-generate-text-sequentially?utm_source=hashnode&utm_medium=feed)

[Pankaj Kumar](https://hashnode.com/@pankajhyd) · [Aug 22, 2025](https://hashnode.com/posts/vision-yatra-step-11-layer-normalization-in-action-step-by-step-math-and-example/68a85c72c9191071a46b7c42)

### [Vision Yatra: Step 11 — Layer Normalization in Action: Step-by-Step Math & Example](https://my-ai-yatra.solveautomation.in/vision-yatra-step-11-layer-normalization-in-action-step-by-step-math-and-example?utm_source=hashnode&utm_medium=feed)

From formula to code — normalize a word vector like a Transformer

4 min read · [my-ai-yatra.solveautomation.in](https://my-ai-yatra.solveautomation.in/vision-yatra-step-11-layer-normalization-in-action-step-by-step-math-and-example?utm_source=hashnode&utm_medium=feed)

[Pankaj Kumar](https://hashnode.com/@pankajhyd) · [Aug 22, 2025](https://hashnode.com/posts/vision-yatra-step-9-positional-encoding-how-transformers-know-word-order/68a855fd58f2f2ebc60c33e7)

### [Vision Yatra: Step 9 — Positional Encoding: How Transformers Know Word Order](https://my-ai-yatra.solveautomation.in/vision-yatra-step-9-positional-encoding-how-transformers-know-word-order?utm_source=hashnode&utm_medium=feed)

Without RNNs, how do Transformers "see" the sequence? The magic of sine and cosine

5 min read · [my-ai-yatra.solveautomation.in](https://my-ai-yatra.solveautomation.in/vision-yatra-step-9-positional-encoding-how-transformers-know-word-order?utm_source=hashnode&utm_medium=feed)

[Pankaj Kumar](https://hashnode.com/@pankajhyd) · [Aug 22, 2025](https://hashnode.com/posts/vision-yatra-step-8-multi-head-attention-to-feed-forward-how-outputs-are-combined/68a85368fd65fd91f63a0534)

### [Vision Yatra: Step 8 — Multi-Head Attention to Feed-Forward: How Outputs Are Combined](https://my-ai-yatra.solveautomation.in/vision-yatra-step-8-multi-head-attention-to-feed-forward-how-outputs-are-combined?utm_source=hashnode&utm_medium=feed)

From 8 attention heads to one vector — the final step before the feed-forward network

5 min read · [my-ai-yatra.solveautomation.in](https://my-ai-yatra.solveautomation.in/vision-yatra-step-8-multi-head-attention-to-feed-forward-how-outputs-are-combined?utm_source=hashnode&utm_medium=feed)

[Pankaj Kumar](https://hashnode.com/@pankajhyd) · [Aug 22, 2025](https://hashnode.com/posts/vision-yatra-step-7-multi-head-attention-how-transformers-learn-multiple-perspectives/68a84c4ac6b608dce543f540)

### [Vision Yatra: Step 7 — Multi-Head Attention: How Transformers Learn Multiple Perspectives](https://my-ai-yatra.solveautomation.in/vision-yatra-step-7-multi-head-attention-how-transformers-learn-multiple-perspectives?utm_source=hashnode&utm_medium=feed)

Go beyond single attention — understand how multiple heads capture syntax, semantics, and context

4 min read · [my-ai-yatra.solveautomation.in](https://my-ai-yatra.solveautomation.in/vision-yatra-step-7-multi-head-attention-how-transformers-learn-multiple-perspectives?utm_source=hashnode&utm_medium=feed)
