Ito Yugo PRO
YUGOROU
AI & ML interests
高校1年 RL･SFT/local LLM･TTS for iOS
Recent Activity
updated a model about 5 hours ago
YUGOROU/lumi-grpo published a model about 5 hours ago
YUGOROU/lumi-grpo updated a dataset 2 days ago
YUGOROU/lumi-dataOrganizations
docs: ルートREADME.mdを追加（YAMLメタデータ・パイプライン全体像）
#14 opened about 2 months ago
by YUGOROU
docs: README.mdを追加（YAMLメタデータ・パイプライン概要）
#15 opened about 2 months ago
by YUGOROU
docs: README.mdにYAMLメタデータを追加
#5 opened about 2 months ago
by YUGOROU
refactor: patchディレクトリを廃止し差分をdata/に統合
#4 opened about 2 months ago
by YUGOROU
refactor: patchディレクトリを廃止し差分をdata/に統合
#2 opened about 2 months ago
by YUGOROU
Fix-SP1: integrate system prompt patch into setup pipeline
1
#13 opened about 2 months ago
by YUGOROU
Fix-SP1: system prompt injection for Base model format guidance
#1 opened about 2 months ago
by YUGOROU
feat: PR#12 - Vast.ai + HF Inference Providers統合・エラーハンドリング改善
#12 opened about 2 months ago
by YUGOROU
feat: PR#11 - EQ-Bench3 macOS評価セットアップ (uv + llama-cpp + HF Inference Providers)
#11 opened about 2 months ago
by YUGOROU
feat: PR#10 - EQ-Bench3 ローカルvLLM評価スクリプト
2
#10 opened about 2 months ago
by YUGOROU
feat: PR#9 - EQ-Bench3 日本語化スクリプト
3
#9 opened about 2 months ago
by YUGOROU
feat: PR#8 - SFT/DPO 学習スクリプト追加
3
#8 opened about 2 months ago
by YUGOROU
fix: PR#7 - chosen/rejectedの<synth>タグ除去
#7 opened about 2 months ago
by YUGOROU
feat: PR#6 - EQ全展開・max-cycles CLI・<think>統一・フィルタ改善（mainから再実装）
1
#6 opened about 2 months ago
by YUGOROU
feat: PR#5 - EQ全展開・max-cycles CLI・<think>統一・フィルタ改善
3
#5 opened about 2 months ago
by YUGOROU
fix: LLMのオブジェクト形式JSON応答に対応
1
#4 opened about 2 months ago
by YUGOROU
fix: enable_thinking制御の分離・content取得ロジック修正
1
#3 opened about 2 months ago
by YUGOROU
fix: マトリックス生成のセマフォ追加・max_tokens拡張
1
#2 opened about 2 months ago
by YUGOROU
fix: マトリックス生成のセマフォ追加・max_tokens拡張
1
#2 opened about 2 months ago
by YUGOROU