Small-compute adaptation plan (Mac-first) - NIC-324#1
Open
Small-compute adaptation plan (Mac-first) - NIC-324#1
Conversation
Implements a deterministic policy engine for evaluating autoresearch results based on validation bits-per-byte (val_bpb) and a complexity score. The core logic is to prioritize candidates with lower val_bpb, while also favoring simplicity. The engine handles crash and timeout statuses explicitly. Includes a comprehensive unit test suite to validate the decision logic under various conditions. Resolves NIC-320.
- Parameter downsizing matrix for DEPTH, batch size, seq len, vocab - TinyStories dataset recommendation for low-entropy training - Throughput/quality envelope for MacBook Pro/Air/CPU configs - Notable fork analysis and implementation strategy - Configuration templates for immediate deployment Addresses NIC-324
- digest.py: Generate HTML reports with run leaderboards and trend charts - log_run.py: Capture experiment results to results.tsv - experiment_runner.py: High-level automation for experiments and digests - Complexity-adjusted scoring: val_bpb + parameter penalty - Ready-for-review summary templates - Integrates with existing policy_engine for autonomous research loops Addresses Linear issue NIC-327: Autoresearch daily research digest and leaderboard
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Comprehensive adaptation plan for running autoresearch on small compute environments, prioritizing MacBooks with Apple Silicon.
Key Deliverables
✅ Parameter downsizing matrix: DEPTH, batch size, sequence length, vocabulary size configurations for MacBook Pro/Air/CPU environments
✅ Dataset recommendations: TinyStories for low-entropy training on resource-constrained setups
✅ Throughput/quality envelope: Expected performance metrics and quality tradeoffs for each platform tier
✅ Implementation strategy: Phase-by-phase rollout starting with MacBook Pro, notable fork analysis
Technical Highlights
Ready for Implementation
Document provides concrete parameter templates and configuration changes ready for immediate deployment.
Closes NIC-324