karpathy/llm.c

init from scratch

Open

#243 opened on Apr 24, 2024

View on GitHub
 (4 comments) (0 reactions) (0 assignees)Cuda (3,626 forks)github user discovery
good first issue

Repository metrics

Stars
 (30,145 stars)
PR merge metrics
 (PR metrics pending)

Description

Follow the GPT-2 reference .py file and initialize the weights in C from scratch in the exact same way. Allow init from scratch instead of init from checkpoint when building the GPT-2. Add argparse flag to configure which way to go. Ok to only change the mainline development file train_gpt2.cu.

Contributor guide

Research direction
Study the reference Python implementation (model.py) to understand weight initialization, then replicate the logic in C in train gpt2.cu, and add an argparse flag to choose between scratch and checkpoint.
Tech stack
c
Domain
backendmachine learningai
Issue type
Feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
2
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
Half day
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Fresh
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
CCUDA
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
60