bingshuo/guanghulab
1
0
Fork 0
Code Issues 2 Pull Requests Actions 1 Packages Projects Releases Wiki Activity

D100 训练状态更新: GPU训练中 loss=2.245

Browse Source
This commit is contained in:
bingshuo 2026-05-18 00:11:36 +08:00
parent c4990152b5
commit dea2601d3a
1 changed files with 20 additions and 25 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

45
brain/training-status.json
View File

@ -1,46 +1,41 @@
{ {
"version": "1.0", "version": "1.0",
"last_updated": "2026-05-17T23:30+08:00", "last_updated": "2026-05-18T00:13+08:00",
"updated_by": "铸渊 (ICE-GL-ZY001)", "updated_by": "铸渊 (ICE-GL-ZY001)",
"mother_model": { "mother_model": {
"status": "tokenizing", "status": "training",
"status_label": "🟡 训练中（分词阶段）", "status_label": "🟢 训练中",
"model": "Qwen/Qwen2.5-7B", "model": "Qwen/Qwen2.5-7B",
"progress": "7,102 / 11,470 条 (62%)", "current_step": 1,
"total_tokens": 21181016, "current_loss": 2.245,
"loss_tokens": 18241745, "grad_norm": 18.88,
"loss_pct": 86.1, "learning_rate": 4.537e-06,
"current_epoch": 0, "current_epoch": 0.035,
"total_epochs": 3, "total_epochs": 3,
"current_step": 0, "gpu": {
"current_loss": null, "util_pct": 97,
"gpu_used_mb": 15257, "mem_used_mb": 77537,
"gpu_total_mb": 97887, "mem_total_mb": 97887,
"gpu_temp_c": 38, "temp_c": 73
"elapsed_hours": 1.0, },
"estimated_remaining_hours": 12.18 "elapsed_hours": 1.2,
"estimated_remaining_hours": 16
}, },
"code_model": { "code_model": {
"status": "pending", "status": "pending",
"status_label": "⚪ 待启动", "status_label": "⚪ 待启动",
"model": "Qwen/Qwen2.5-Coder-7B", "model": "Qwen/Qwen2.5-Coder-7B"
"trigger": "母模型训练完成 → 自动上传COS → 自动启动"
}, },
"pipeline": { "pipeline": {
"status": "running",
"steps": [ "steps": [
{"name": "母模型训练", "status": "in_progress", "label": "🟡 进行中"}, {"name": "母模型训练", "status": "training", "label": "🟢 训练中 (loss=2.245)"},
{"name": "上传COS (母模型)", "status": "pending", "label": "⚪ 等待中"}, {"name": "上传COS (母模型)", "status": "pending", "label": "⚪ 等待中"},
{"name": "代码模型训练", "status": "pending", "label": "⚪ 等待中"}, {"name": "代码模型训练", "status": "pending", "label": "⚪ 等待中"},
{"name": "上传COS (代码模型)", "status": "pending", "label": "⚪ 等待中"} {"name": "上传COS (代码模型)", "status": "pending", "label": "⚪ 等待中"}
] ]
}, },
"alerts": [ "alerts": [],
{"level": "info", "message": "分词完成约需5分钟，之后进入GPU训练"}
],
"links": { "links": {
"notion": "https://www.notion.so/363fb92f383181029fafef8438d6c300", "notion": "https://www.notion.so/363fb92f383181029fafef8438d6c300"
"cognitive_chain": "brain/d100-cognitive-chain.md",
"session_record": "brain/d100-session-record.md"
} }
} }