Modelpublic

agentic-ptb/dpsk-v4-flash.h082.sft5.step_800

sourceHugging Faceupdated 26d agoView on Hugging Face

0likes21downloads

Model Card

dpsk-v4-flash.sft5.step_800

AgentPTB sweep checkpoint. Cell `dpsk-v4-flash` — pi / DeepSeek v4-flash @ effort `thinking`.

field	value
plot cell	`dpsk-v4-flash`
driver	pi / DeepSeek v4-flash
reasoning effort	`thinking`
run boot (UTC)	2026-08-11T08:38:35Z
role	intermediate
origin	`msr-spare/msr-agentic-ptb-dpsk-sft5-intermediates` :: `step_800`
shards	1
base model	`Qwen/Qwen3.5-9B-Base`
eostokenid	`[248044, 248046]` ✅ correct (repaired 2026-08-28)

The local copy of this checkpoint was pruned from the PVC; it was recovered from the msr-spare backup listed under origin.

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.