novastar111/pacman_hard_noncot_chunk_k3_train
pacman_hard_noncot_chunk_k3_train BAGEL VLM-Gym world-model dataset (pacman / noncot). Non-CoT chunk-K train set (no imagined reasoning); re-grounds every K=3 steps. layout: Train-only. Gzipped-JSONL shards under training/; each row is one packed SFT sample with base64-JPEG frames inline. images are base64-encoded JPEG frames stored inline in each JSONL row. Pairs with the matching pacman checkpoint(s) under the companion model org; CoT and non-CoT variants share the same… See the full description on the dataset page: https://huggingface.co/datasets/novastar111/pacman_hard_noncot_chunk_k3_train.
pacmanhardnoncotchunkk3_train
BAGEL VLM-Gym world-model dataset (pacman / noncot).
- Non-CoT chunk-K train set (no imagined reasoning); re-grounds every K=3 steps.
- layout: Train-only. Gzipped-JSONL shards under
training/; each row is one packed SFT sample with base64-JPEG frames inline. - images are base64-encoded JPEG frames stored inline in each JSONL row.
Pairs with the matching pacman checkpoint(s) under the companion model org; CoT and non-CoT variants share the same underlying rollouts.
