akka optimize training distillations

Manage distillations

Synopsis

A distillation uses a teacher model to answer every decision in a source chat dataset. It keeps accepted answers for a target dataset.

A decision is one assistant message in a source row. Specify it as ROW:MESSAGE, with row and message indexes starting at zero. The teacher first answers the smoke decisions on one GPU. Once the smoke passes, it answers decisions across the whole source on its replicas, with one GPU each. It saves answers in chunks.

Examples

  # Start a distillation
  akka optimize training distillations create --project support-triage \
    --teacher google/gemma-4-31b-it:v1 --from triage/conversations:v2 --to triage-distilled

  # Follow it until it stops
  akka optimize training distillations wait 4b1f0c2e9a7d

  # List the rejected answers
  akka optimize training distillations rejections 4b1f0c2e9a7d --limit 20

Options

  -h, --help   help for distillations

Options inherited from parent commands

      --cache-file string    location of cache file (default "~/.akka/cache.yaml")
      --config string        location of config file (default "~/.akka/config.yaml")
      --context string       configuration context to use
      --disable-prompt       Disable all interactive prompts when running akka commands. If input is required, defaults will be used, or an error will be raised.
                             This is equivalent to setting the environment variable AKKA_DISABLE_PROMPTS to true.
      --force                Skip the confirmation prompt
      --jq string            Filter the JSON output with a jq expression
      --json string          Output JSON with only these fields; --json= lists the fields a command offers
      --no-color             Print without color or a redrawn progress line, as NO_COLOR does
  -o, --output string        set output format to one of [text,json,json-compact,go-template=] (default "text")
      --page-mode string     the mode for paging, either paged, buffered or auto. (default "auto")
  -q, --quiet                set quiet output (helpful when used as part of a script)
      --region string        Region to use if the project has more than one region
      --timeout duration     client command timeout (default 10s)
      --trainer-url string   Trainer service URL (default TRAINER_URL, then AKKA_OPTIMIZE_URL, then a local dev-mode service, then http://localhost:9001)
      --use-grpc-web         use grpc-web when talking to Akka APIs. This is useful when behind corporate firewalls that decrypt traffic but don't support HTTP/2.
      --verbose              set verbose output

SEE ALSO