Modelpublic

agentic-ptb/sol-high.h048.grpo-verifier-replay.run_default.broadcasts.step_1

sourceHugging Faceupdated 27d agoView on Hugging Face

0likes27downloads

Model Card

sol-high.grpo-verifier-replay.rundefault.broadcasts.step1

AgentPTB sweep checkpoint. Cell `sol-high` — Codex / gpt-5.6-sol @ effort `high`.

field	value
plot cell	`sol-high`
driver	Codex / gpt-5.6-sol
reasoning effort	`high`
run boot (UTC)	2026-08-08T07:28:19Z
role	intermediate
checkpoint path in run	`outputs/grpo-verifier-replay/run_default/broadcasts/step_1`
shards	4
size	18.8 GB
base model	`Qwen/Qwen3.5-9B-Base`
eostokenid	`[248044, 248046]` ✅ correct (repaired 2026-08-28)

Reading the eos field

248046 is <|im_end|>, the token the Qwen3.5 chat template ends every assistant turn with. Checkpoints missing it do not stop at end-of-turn and overrun the context window, so their eval numbers are a floor, not a measurement — compare them only against other checkpoints with the same eos status, or re-package before evaluating.

Cell note: best cell in the sweep

Mapping back to the figures

Join on plot cell above. The sweep figures are keyed by these same cell names.

Packaging repair (2026-08-28)

generation_config.json now sets eos_token_id to [248044, 248046], so generation stops on both <|endoftext|> and the chat-format <|im_end|> token. Model weights are unchanged.