Demo data: three made-up repos and three people, recorded by the real Trail hooks.About TrailGitHub
Trail/ml-pipeline
ml-pipelineDemo
LleoStop training early when eval loss plateaus06c5771· 11 hours ago 5 Commits
index5 files
users30 files
repo.json113 Bytes
Recent agent sessions
Runs keep going long after eval loss stops improving. Add early stopping.claude11 hours ago
Eval numbers look too good. Check whether eval prompts leak into train and fix the split.codex3 days ago
The 7B LoRA fine-tune keeps blowing up around step 400. Logs are in logs/run-17.log. Fix it.claude6 days ago

About

Agent sessions, logs and artifacts linked to every commit of ml-pipeline.

acme/ml-pipeline (demo, not on GitHub)
5 commits
3 agent sessions
7 artifacts · 38.5 KB
Updated 11 hours ago

Contributors 3

LMS
leo@acme.dev
maya@acme.dev
sam@acme.dev