tinker_cookbook.distillation.Config
class tinker_cookbook.distillation.Config(**)
Configuration for SDFT training.
Fields:
- model_name (str) – Model
- renderer_name (str | None) – Default:
None - lora_rank (int) – Default:
128 - base_url (str | None) – Default:
None - learning_rate (float) – Training Default:
2e-05. - max_tokens (int) – Default:
2048 - temperature (float) – Default:
1.0 - loss_fn (LossFnType) – Default:
'cross_entropy' - topk (int) – SDFT-specific Default:
20. - reverse (bool) – Default:
False - demo_template (str) – Default:
DEFAULT_DEMO_TEMPLATE - system_prompt (str | None) – Default:
None - teacher_sync_every (int | None) – Default:
None - max_context_length (int) – Default:
32768 - evaluator_builders (list[SamplingClientEvaluatorBuilder]) – Evaluation Default:
[]. - eval_every (int) – Default:
20 - save_every (int) – Default:
20 - num_substeps (int) – Standard infra Default:
1. - log_path (str)
- wandb_project (str | None) – Default:
None - wandb_name (str | None) – Default:
None - load_checkpoint_path (str | None) – Default:
None - max_steps (int | None) – Default:
None - enable_trace (bool) – Default:
False - span_chart_every (int) – Default:
0