 David Gil
|
47b705409e
fix: BM25 score normalization - use Math.abs instead of Math.max (#76)
|
3 kuukautta sitten |
|
Christopher Stöckl
|
0f87e2429d
fix: workaround Bun UTF-8 path corruption bug (#82)
|
3 kuukautta sitten |
|
Matthías Páll Gissurarson
|
5de063ae96
Fix: Add missing --index option to argument parser (#84)
|
3 kuukautta sitten |
|
Tobi Lütke
|
102ff861d3
fix: use Qwen3 recommended sampling params to prevent repetition loops
|
3 kuukautta sitten |
|
Tobi Lütke
|
479b68bbf1
add qmd model pull and refresh logic
|
3 kuukautta sitten |
|
Tobi Lütke
|
bf1b8fc90a
lots of training stuff
|
3 kuukautta sitten |
|
Tobi Lutke
|
17c201ea81
fix: correct QMD acronym to Query Markup Documents
|
3 kuukautta sitten |
|
Tobi Lutke
|
739038e1a7
docs: add explicit HuggingFace repo destinations
|
3 kuukautta sitten |
|
Tobi Lutke
|
38073799c0
chore: clean up finetune folder and fix training workflow
|
3 kuukautta sitten |
|
Tobi Lutke
|
533f0eed37
docs: add finetune CLAUDE.md and update training workflow
|
3 kuukautta sitten |
|
Tobi Lutke
|
7de18ee066
Merge main into finetune
|
3 kuukautta sitten |
|
Tobi Lutke
|
785620467a
refactor: reorder output format to put hyde line first
|
3 kuukautta sitten |
|
Tobi Lütke
|
8cc7d8c138
Add sampled /only: variants (399) for training balance
|
3 kuukautta sitten |
|
Tobi Lütke
|
20aef8a3e9
Change format to /only:lex (slash prefix)
|
3 kuukautta sitten |
|
Tobi Lütke
|
46ff098361
Change only: format to only:lex (no space after colon)
|
3 kuukautta sitten |
|
Tobi Lütke
|
806a0cfc14
Add 'only:' mode support for single-type expansions
|
3 kuukautta sitten |
|
Tobi Lütke
|
32d313ad6b
Add LLM session management for lifecycle safety
|
3 kuukautta sitten |
|
Christopher Jones
|
6d9871d2f5
Fix DisposedError during slow batch embedding (#41)
|
3 kuukautta sitten |
|
Algimantas Krasauskas
|
f6a987a642
Add skills.sh integration for AI agent discovery (#64)
|
3 kuukautta sitten |
|
Tobi Lutke
|
7b98d4d308
Link fine-tuned model to HuggingFace in README
|
3 kuukautta sitten |
|
Tobi Lutke
|
5cf4958bfa
Add HuggingFace model card YAML metadata to finetune README
|
3 kuukautta sitten |
|
Tobias Lütke
|
eb1b77c8cb
Deploy fine-tuned GRPO model as default query expansion (#67)
|
3 kuukautta sitten |
|
Tobi Lutke
|
8572c2fd94
Deploy fine-tuned GRPO model as default for query expansion
|
3 kuukautta sitten |
|
Tobi Lutke
|
5ab78d00a2
Add HF Jobs scripts, temporal query examples, and training results
|
3 kuukautta sitten |
|
Tobi Lutke
|
354744af53
Finetune 2.0: consolidate and simplify the entire training pipeline
|
3 kuukautta sitten |
|
jdvmi00
|
64c6e6c2e3
fix: rename collectionId to collectionName in searchVec for proper filtering (#61)
|
3 kuukautta sitten |
|
Freeman Jiang
|
bfb0eebc3e
fix: use sequential embedding on CPU-only systems to avoid race condition (#54)
|
3 kuukautta sitten |
|
Copilot
|
053252ca24
Add Windows path utilities with cross-platform test coverage (#51)
|
3 kuukautta sitten |
|
sh54
|
ba7391832d
Add org-mode title extraction support (#50)
|
3 kuukautta sitten |
|
sh54
|
65c0f89560
Enable SQLite extension loading in devshell (#48)
|
3 kuukautta sitten |
