Observability

These commands read what runs recorded: rows, events, traces, and the difference between two runs. Metrics and Traces explain the output.

Command	Description
`monte run list`	List the runs of an experiment
`monte run show`	Show one run in full
`monte run logs`	Print or follow the events of a run
`monte run stop`	Stop a running run
`monte trace list`	List the traces of a run
`monte trace download`	Write the traces of a run to a file
`monte compare`	Compare two eval runs task by task

monte run list

monte run list <experiment> [--kind eval|train] [--status <status>] [--all] [--json]

List the runs of an experiment, oldest first, with the alias of each succeeded run (eval@200), its status, and its score as <eval>=<score>±<interval>.

Flag	Type	Default	Description
`<experiment>`	argument	required	The experiment
`--kind`	string	both	`eval` or `train`
`--status`	string	all	`running`, `succeeded`, `failed`, or `abandoned`
`--all`	flag	off	Include smoke runs, marked `[not evidence]`
`--json`	flag	off	Print JSON on stdout

monte run list support-triage

RUN       ALIAS       STATUS     DETAIL                                        SCORE
eval-01   eval@0      succeeded  support-escalation-dev  base                  support-escalation-dev=0.50±0.07
eval-02   eval@0      succeeded  support-escalation      base                  support-escalation=0.49±0.07
...
train-07  train@200   succeeded  triage-sft  support-replies  from base
eval-03   eval@200    succeeded  support-escalation-dev  train-07/200          support-escalation-dev=0.60±0.07
train-09  -           running    triage-sft  support-replies  from train-07/200

monte run show

monte run show <experiment>/<run> [--metrics <tags>] [--json]

Show one run in full: status, target, settings, scores, eval checks, and provenance. A training run also shows its training config, training data version, and checkpoint.

Flag	Type	Default	Description
`<experiment>/<run>`	argument	required	The run, by ID or alias
`--metrics`	string	all	With `--json`, include the per-step series for these comma-separated metric tags
`--json`	flag	off	Print JSON on stdout

monte run show support-triage/eval-03

run         support-triage/eval-03 (eval)  succeeded
target      train-07/200 on Qwen/Qwen3-8B @ a09a3545
score       0.60 ± 0.07
vs baseline +0.10 ± 0.06 paired (30 fixed / 10 broken, eval-01)
checks      mean_output_tokens 142  parse_fail_rate 0.00  truncations 0.01  repetition 0.02

monte run logs

monte run logs <experiment>/<run> [-f] [--tail <n>]

Print the events of a run as JSON lines: progress, checkpoint_saved, and score_recorded, and a final status event.

Flag	Type	Default	Description
`<experiment>/<run>`	argument	required	The run
`-f`	flag	off	Keep the output open and print new events until the run ends
`--tail`	integer	all	Print only the last `n` events first

monte run logs support-triage/train-07 -f --tail 2

{"event": "progress", "step": 200, "loss": 0.407, "step_time_s": 4.0}
{"event": "checkpoint_saved", "checkpoint": "train-07/200", "kind": "both"}
{"event": "status", "status": "succeeded"}

monte run stop

monte run stop <experiment>/<run> [--json]

Stop a running run. Its status becomes abandoned. Run history keeps what the run recorded, and a training run keeps the checkpoints it saved.

Flag	Type	Default	Description
`<experiment>/<run>`	argument	required	The run
`--json`	flag	off	Print JSON on stdout

monte run stop support-triage/train-09

Stopped support-triage/train-09 at step 264 of 300. Status abandoned. Last checkpoint train-09/250.

Error	Cause
`E_NOT_RUNNING`	The run is not running

monte trace list

monte trace list <experiment>/<run> [--eval <eval>] [--failed] [--step <n>] [--sort task|score] [--json]

List the traces of a run, one row for each task, with the score, the parsed and expected answers, the tokens, and the stop reason. A training run with a grader has one trace for each scored rollout.

Flag	Type	Default	Description
`<experiment>/<run>`	argument	required	The run
`--eval`	string	the eval that the run scored	The eval, for a run with traces on more than one
`--failed`	flag	off	Only the traces with a score below `1.0`
`--step`	integer	all steps	For a training run, only the rollouts of one step
`--sort`	string	`task`	`task` or `score`
`--json`	flag	off	Print JSON on stdout

monte trace list support-triage/eval-03 --failed --sort score

TASK    SCORE  EXTRACTED                 EXPECTED                 TOKENS     STOP
t-004   0.00   resolve                   escalate: money          412/96     -
t-031   0.00   escalate: angry           escalate: legal          388/41     -
t-142   0.50   resolve                   resolve                  520/400    length
79 failed of 200

monte trace download

monte trace download <experiment>/<run> [--eval <eval>] [--step <n>] [-o <path>] [--json]

Write the traces of a run to a JSONL file, one trace on each line, with the fields that Trace Contents lists.

Flag	Type	Default	Description
`<experiment>/<run>`	argument	required	The run
`--eval`	string	the eval that the run scored	The eval
`--step`	integer	all steps	For a training run, only the rollouts of one step
`-o`	path	`./<experiment>-<run>-traces.jsonl`	The output file. `-` writes to stdout.
`--json`	flag	off	Print a JSON summary on stdout

monte trace download support-triage/eval-03 -o eval-03-traces.jsonl

Wrote 200 traces (3.4 MB) to eval-03-traces.jsonl.

monte compare

monte compare <experiment>/<run> <experiment>/<run> [--show fixed|broken|same] [--json]

Compare two eval runs on the same eval, task by task. The first run is the reference. Compare Runs explains the output.

Flag	Type	Default	Description
`<experiment>/<run>`	argument	required	The reference run
`<experiment>/<run>`	argument	required	The compared run
`--show`	string	none	Also list the task IDs of one group
`--json`	flag	off	Print JSON on stdout

monte compare support-triage/eval@0 support-triage/eval-03 --show broken

difference  +0.10 ± 0.06 paired   30 fixed / 10 broken / 160 same
broken      t-004 t-019 t-031 t-052 t-077 t-090 t-128 t-144 t-171 t-199

Error	Cause
`E_NOT_COMPARABLE`	The runs scored other evals, belong to other experiments, have other base models, or one is a smoke run

Commands

MCP

monte run list

monte run show

monte run logs

monte run stop

monte trace list

monte trace download

monte compare

​monte run list

​monte run show

​monte run logs

​monte run stop

​monte trace list

​monte trace download

​monte compare

monte run list

monte run show

monte run logs

monte run stop

monte trace list

monte trace download

monte compare