No local training runs
Set
report_to=simpletuner in a training environment to record one.
Status
Model family
Latest step
Samples
Scalars
Validation
Compare saved outputs by prompt and step.No validation media was recorded for this run.
{% call hero_cta(
theme='indigo',
icon_class='fa-chart-line',
icon_shape='circle',
title='Metrics & Observability',
description='Export metrics to Prometheus for monitoring dashboards. Track jobs, costs, health, and security events in Grafana, Datadog, or any Prometheus-compatible system.',
features=[
{'icon': 'fa-tasks', 'color': 'text-primary', 'label': 'Job & Cost Tracking'},
{'icon': 'fa-heartbeat', 'color': 'text-success', 'label': 'Health Monitoring'},
{'icon': 'fa-shield-alt', 'color': 'text-warning', 'label': 'Security Metrics'},
{'icon': 'fa-plug', 'color': 'text-info', 'label': 'Grafana Ready'},
],
show_condition='true',
dismiss_method='dismissHero()',
tip_text='Metrics are available at
/api/metrics/prometheus'
) %}
{{ template_buttons([
{'label': 'Minimal', 'icon': 'fa-feather', 'action': "applyTemplate('minimal')", 'title': 'Just job counts - lightweight monitoring'},
{'label': 'Standard', 'icon': 'fa-check-circle', 'action': "applyTemplate('standard')", 'title': 'Jobs, HTTP metrics, and health checks', 'class': 'btn-primary'},
{'label': 'Security', 'icon': 'fa-lock', 'action': "applyTemplate('security')", 'title': 'Include rate limiting, audit, and approvals'},
{'label': 'Full', 'icon': 'fa-th', 'action': "applyTemplate('full')", 'title': 'All available metrics'},
]) }}
{{ integration_card(
icon='fa-chart-bar',
title='Grafana',
description='Create dashboards with job metrics, cost tracking, and health status'
) }}
{{ integration_card(
icon='fa-bell',
title='Alerting',
description='Set up alerts for job failures, cost limits, and rate limit violations'
) }}
{% endcall %}
Circuit Breakers
All services healthy - no circuit breakers active
Circuit breakers protect against cascading failures
GPU Health
No GPU information available
๐ฅ
VRAM
Utilization
Fan
Temperature
History
Prometheus Export
/api/metrics/prometheus
metrics,
KB,
ms
Add to Prometheus
scrape_configs
Export Preview
Click refresh to load preview
Loading metrics...
Prometheus Config
scrape_configs:
- job_name: 'simpletuner'
static_configs:
- targets: ['localhost:8080']
metrics_path: '/api/metrics/prometheus'
scrape_interval: 30s
Copied to clipboard!