56,966 blogs · [ { "id": "01a0e1c8-1d95-73f9-bb0c-193eef273c71", "title": "Policy Gradient for LLMs, Explained Visually", "url": "https://tylerromero.com/posts/2026-09-policy-gradient/", "published_at": "2026-09-27T07:00:00+00:00" }, { "id": "01a08c59-e015-713c-8022-885820f13132", "title": "NanoGPT Speedrun Worklog", "url": "https://tylerromero.com/posts/nanogpt-speedrun-worklog/", "published_at": "2025-03-08T08:00:00+00:00" }, { "id": "01a08c59-e015-713c-8022-885821d000ba", "title": "Reducing VRAM Footprint in PPO and GRPO Using Selective Log-Softmax", "url": "https://tylerromero.com/posts/2025-02-selective-log-softmax/", "published_at": "2025-02-06T08:00:00+00:00" }, { "id": "01a08c59-e015-713c-8022-8858222650e7", "title": "An Extension to BADGE Active Learning for Variable-Sized Batches", "url": "https://tylerromero.com/posts/2025-01-badge-extension/", "published_at": "2025-01-05T08:00:00+00:00" }, { "id": "01a08c59-e015-713c-8022-8858223d86e8", "title": "Direct Preference Optimization Explained In-depth", "url": "https://tylerromero.com/posts/2024-04-dpo/", "published_at": "2024-04-13T08:00:00+00:00" } ] posts Claim your blog

Blog · corpus.blog/blogs/tylerromero.com/posts

tylerromero.com

2026

Policy Gradient for LLMs, Explained Visually original ↗

27 Sept 2026

2025

NanoGPT Speedrun Worklog original ↗

8 Mar 2025

Reducing VRAM Footprint in PPO and GRPO Using Selective Log-Softmax original ↗

6 Feb 2025

An Extension to BADGE Active Learning for Variable-Sized Batches original ↗

5 Jan 2025

2024

Direct Preference Optimization Explained In-depth original ↗

13 Apr 2024