Skip to content
Permalink

Comparing changes

Choose two branches to see what’s changed or to start a new pull request. If you need to, you can also or learn more about diff comparisons.

Open a pull request

Create a new pull request by comparing changes across two branches. If you need to, you can also . Learn more about diff comparisons here.
base repository: johnmarktaylor91/torchlens
Failed to load repositories. Confirm that selected base ref is valid, then try again.
Loading
base: v0.20.4
Choose a base ref
...
head repository: johnmarktaylor91/torchlens
Failed to load repositories. Confirm that selected head ref is valid, then try again.
Loading
compare: v0.20.5
Choose a head ref
  • 3 commits
  • 5 files changed
  • 3 contributors

Commits on Mar 9, 2026

  1. fix(vis): prevent OOM kill on 1M-node ELK render (#128) 

    The 1M-node render was OOM-killed at ~74GB RSS because:
1. Model params (~8-10GB) stayed alive during ELK subprocess
2. preexec_fn forced fork+exec, COW-doubling the 74GB process
3. Heap/stack formulas produced absurd values (5.6TB heap, 15GB stack)
4. No memory cleanup before subprocess launch

Changes:
- render_large_graph.py: separate log_forward_pass from render_graph,
  free model/autograd before ELK render
- elk_layout.py: cap heap at 64GB, stack floor 4096MB/cap 8192MB,
  write JSON to temp file (free string before subprocess), gc.collect
  before subprocess, set RLIMIT_STACK at module level (removes
  preexec_fn and the forced fork+exec)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
    @johnmarktaylor91 @claude
    johnmarktaylor91 and claude committed Mar 9, 2026
    Configuration menu
    Copy the full SHA
    d9a1525 View commit details
    Browse the repository at this point in the history

  2. Merge pull request #128 from johnmarktaylor91/fix/elk-oom-1m-nodes

    @johnmarktaylor91
    johnmarktaylor91 authored Mar 9, 2026
    Configuration menu
    Copy the full SHA
    ef80cfe View commit details
    Browse the repository at this point in the history

  3. chore(release): 0.20.5

    semantic-release committed Mar 9, 2026
    Configuration menu
    Copy the full SHA
    9b65063 View commit details
    Browse the repository at this point in the history
Loading