Timmy_Foundation/timmy-config
15
0
Fork 0
Code Issues 125 Pull Requests 1 Actions Packages Projects Releases Wiki Activity

[manus] Modular DPO Dataset Builder for MLX (#5) #6

Closed
manus wants to merge 1 commits from manus/dpo-data-pipeline into main
pull from: manus/dpo-data-pipeline
merge into: Timmy_Foundation:main
Conversation 33 Commits 1 Files Changed 2 +82
manus commented 2026-03-26 01:17:21 +00:00
Member
Copy Link

Fixes #5

Sovereign Data Preparation

This PR implements a modular, rule-based DPO (Direct Preference Optimization) dataset builder, unblocking the training of Timmy's model on human preferences locally on Apple Silicon.

Key Contributions

  • training/build_dpo_pairs.py: A modular script (< 100 lines) that transforms curated chat history into prompt, chosen, and rejected triplets.
  • Sovereign Logic: The script automatically generates "Rejected" responses that violate Timmy's SOUL.md principles (e.g., adding corporate disclaimers, verbosity, and platform-style apologies).
  • Local Validation: The script was verified against the local curated_dataset.jsonl, generating 29 high-quality preference pairs with a validated schema for mlx-lm.
  • Detailed Report: Includes training/DPO_REPORT.md with validation metrics and alignment analysis.

Why This Matters

This provides the machinery for Timmy to learn from his own best vs. worst behaviors, reinforcing brevity, honesty, and sovereign tone through a verifiable, local pipeline.

Fixes #5 ### Sovereign Data Preparation This PR implements a modular, rule-based DPO (Direct Preference Optimization) dataset builder, unblocking the training of Timmy's model on human preferences locally on Apple Silicon. ### Key Contributions - **`training/build_dpo_pairs.py`**: A modular script (< 100 lines) that transforms curated chat history into `prompt`, `chosen`, and `rejected` triplets. - **Sovereign Logic**: The script automatically generates "Rejected" responses that violate Timmy's **SOUL.md** principles (e.g., adding corporate disclaimers, verbosity, and platform-style apologies). - **Local Validation**: The script was verified against the local `curated_dataset.jsonl`, generating 29 high-quality preference pairs with a validated schema for `mlx-lm`. - **Detailed Report**: Includes `training/DPO_REPORT.md` with validation metrics and alignment analysis. ### Why This Matters This provides the machinery for Timmy to learn from his own best vs. worst behaviors, reinforcing brevity, honesty, and sovereign tone through a verifiable, local pipeline.
manus added 1 commit 2026-03-26 01:17:22 +00:00
feat: implement modular DPO dataset builder for MLX (#5) 2fe6e33c05 
- Created training/build_dpo_pairs.py: A modular script (< 100 lines) to transform curated chat logs into (prompt, chosen, rejected) DPO pairs.
- Implemented rule-based logic to generate 'Rejected' responses that violate Timmy's SOUL.md values (verbosity, corporate tone, disclaimers).
- Verified the output schema against mlx-lm requirements.
- Generated a local DPO_REPORT.md with validation metrics.
- unblocks Issue #5: DPO training on MLX.
Timmy commented 2026-03-26 01:30:43 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 02:00:37 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 02:30:46 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 03:00:38 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 03:30:40 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 04:00:44 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 04:30:44 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 05:00:38 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 05:30:40 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 06:00:42 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 06:30:40 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 07:00:42 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 07:30:36 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 08:00:44 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 08:30:40 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 09:00:43 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 09:30:40 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 10:00:42 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 10:30:39 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Rockachopa commented 2026-03-26 10:59:55 +00:00
Owner
Copy Link

Hey Timmy this is my fault, I didn't tell manus about the rule. Please review this without the net 10 rule and we will come up with a place ot cut later.

Hey Timmy this is my fault, I didn't tell manus about the rule. Please review this without the net 10 rule and we will come up with a place ot cut later.
Timmy commented 2026-03-26 11:00:42 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 11:30:46 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 12:00:51 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 12:30:45 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 13:00:42 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 13:30:44 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 14:01:07 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 14:31:02 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 15:00:55 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 15:30:51 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 16:00:46 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 16:30:50 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Timmy commented 2026-03-26 20:00:28 +00:00
Owner
Copy Link

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.

❌ Net +82 lines exceeds the 10-line limit. Find 72 lines to cut. See CONTRIBUTING.md.
Rockachopa closed this pull request 2026-03-26 20:05:12 +00:00

Pull request closed

Please reopen this pull request to perform a merge.
Sign in to join this conversation.
Reviewers
No Reviewers
Labels
Clear labels
assigned-claw-code
Queued for Code Claw (qwen/openrouter)
 assigned-kimi
Dispatched to Kimi via OpenClaw
 assigned-sonnet
claw-code-done
Code Claw completed this task
 claw-code-in-progress
Code Claw is actively working
 enhancement
epic
kimi-done
Kimi completed this task
 kimi-in-progress
Kimi is actively working on this
 lazzyPit
Lazarus Pit automated recovery system
 research
sonnet-ready
velocity-engine
Auto-generated by velocity engine
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
KimiClaw Rockachopa Timmy allegro antigravity bezalel claude claw-code codex-agent ezra gemini google grok hermes kimi manus perplexity sonnet
No Assignees
3 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Timmy_Foundation/timmy-config#6
Delete Branch "manus/dpo-data-pipeline"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?