vishal

50:20

Understanding ColBERT's ivf.pid.pt: Inspecting Intermediate Artifacts from _build_ivf & optimize_ivf

19:13

Do RAGatouille and ColBERT Produce the Same Index and Retrieval Scores? A Deep Dive Comparison

19:12

TIL: Understanding LLM Foundry's BinPackCollator (Sequence Packing for 95% Token Efficiency!)

23:06

Improving LLM Judge Alignment: Enhancing TinyScale Lab Evaluation Agreement to 94%

26:38

Technical Report Summary: Nomic Embed

21:17

Understanding Sequence Packing: Initial Musings

15:06

Building an LLM Judge Agreement App: 7 Iterations from Basic to Full Functionality

44:00

Evaluating First Attempt LLM Judge Scores: Improving Claude Haiku Alignment for Story Scoring

36:39

Manual Scoring Results for TinyStories Models: Grammar, Reasoning, and Emergent Capabilities

40:23

Look at Your Data: Building an LM Scoring App with FastHTML

30:15

TSL: Curating Evaluation Prompts, Defining Scoring Criteria + Designing LLM Judge Prompt Template

19:41

TinyScale Lab Update: Setting Eval Targets + Generation Completions for LLM Judge Development

9:24

TinyScaleLab Project Update: Training Cost Analysis and Evaluation Infrastructure Plans

26:51

TinyScale Lab: Exploring the Connection Between Training Dynamics and Model Capabilities

1:04:03

Research Paper Summary: Small-scale proxies for large-scale Transformer training instabilities

22:47

My Second-Place Winning Tiny Model Hackathon Journey: Pre-Training from Scratch

21:19

LossInspector: A Deep Dive Into LLM-Foundry's Next-Token Prediction with a Custom Composer Callback

28:41

The Evolution of Matrix Multiplication: 12,000x Numba Speedup 🚀 | fastai Course Lesson 11

1:23:34

Research Paper Summary: TinyStories

47:38

Look at Your Data: Manual Validation of Retrieval Metrics

46:10

Creating a Custom Composer Callback to Track Data Types in LLM Training | Mixed Precision Deep Dive

1:16:36

Paper Reading: Small-scale proxies for large-scale Transformer training instabilities

1:36:58

Paper Reading: Overtrained Language Models Are Harder to Fine-Tune

55:01

Paper Reading: SmolLM2

20:16

Exploring Sequential and Merged Linear Layer Forward Passes

9:21

TIL: Using PyTorch's register_forward_hook to Trace Floating Point Errors

26:21

Debugging Un-Merged and Merged LoRA Model Output Differences

18:10

LoraModel.merge_and_unload Deep Dive

1:05:43

RAGatouille/ColBERT Indexing Deep Dive

25:43

Recreating Plots from Appendix A.4 of the DoRA Paper for LoRA Learns Less and Forgets Less Models

17:43

TIL: PeftModel Base Model Behavior

41:51

Research Paper Summary: Hypencoder: Hypernetworks for Information Retrieval

26:53

Code Walkthrough - peft DoRA Implementation

19:42

Research Paper Summary: rsLoRA

36:57

Research Paper Summary: LoRA Learns Less and Forgets Less

1:14:59

Recreating the PLAID ColBERTv2 Scoring Pipeline: From Research Code to RAGatouille

11:48

fastbook-benchmark: ColBERT Search

18:09

fastbook-benchmark: Single Vector Search

20:36

fastbook-benchmark: Scoring Retrieval Results

16:37

fastbook-benchmark: Full Text Search Implementation

17:01

fastbook-benchmark: Document Processing

4:27

Introducing the fastbook-benchmark Information Retrieval QA Dataset

21:27

Implementing Image to Image Generation in Stable Diffusion from Scratch | fastai Part 2

37:34

Implementing Negative Prompting in Stable Diffusion from Scratch | fastai Part 2

38:55

fastai - Chapter 8 - Collaborative Filtering Deep Dive Code Walkthrough

25:12

Implementing a Custom Test Time Augmentation Method using fastai

2:41:31

fastai - Chapter 6 - Building Single, Multi-label Classification and Image Regression Models