alt-glitch 0ab46d65b0 Add streaming eval sample logging to BaseEnv ... Introduces `log_eval_sample()` method for stream-writing individual evaluation samples to `samples.jsonl` during evaluation, with lazy writer initialization and automatic HTML generation on completion. Updates GSM8k environment to use streaming approach instead of batching samples.		2026-03-28 00:52:13 -07:00
..
api	clean log	2026-03-13 12:12:33 -04:00
cli	fix: pass num_steps to register_to_api	2026-02-24 11:22:18 +02:00
envs	Add streaming eval sample logging to BaseEnv	2026-03-28 00:52:13 -07:00
frontend	add trajectory saving to eval mode.	2026-03-27 16:04:04 -07:00
tests	revert teacher logprobs	2026-03-23 11:23:47 -07:00
utils	fix duplicate code + add safety checks	2026-02-09 10:58:49 +03:30
__init__.py	first commit	2025-04-29 12:10:10 -07:00
FAQ.md	Update FAQ.md	2026-01-29 10:43:47 +02:00
type_definitions.py	Update type_definitions.py	2025-10-23 10:26:51 +02:00