datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
behavior-1k-2025-challenge-vjepa2-vitg-demo-embeddings
V-JEPA 2 ViT-G Embeddings — BEHAVIOR-1K 2025 Challenge Demos (62h)
Precomputed video embeddings for a 62-hour subsample of the
BEHAVIOR-1K 2025 challenge demonstrations,
extracted with the V-JEPA 2 ViT-g encoder.
The goal is to make downstream experimentation faster and more reproducible by eliminating
repeated video decoding and encoder forward passes — lowering the barrier for teams
without access to large GPU clusters.
Field
Value
Source dataset… See the full description on the dataset page: https://huggingface.co/datasets/quastAI/behavior-1k-2025-challenge-vjepa2-vitg-demo-embeddings.bridgev2_vjepa21_latent_shardedsokoban-10k-vjepa2-tokenizedlewm-vjepa21-state-moe-shared-residual-wo-tokencl-analysis
LEWM V-JEPA2.1 State-MoE Shared-Residual Analysis
This dataset contains PNG visualizations and adjacent-epoch absolute-delta
summaries for the experiment
lewm_reasoning_vjepa21_vitL_tokens_100tasks_stateMoE_sharedResidual_pair_topkExcess_flat_headaware_woTokenCL_modify.
Contents
metadata.csv: parsed stage, epoch, layer, condition, scope, and image paths.
gallery_manifest.json: manifest consumed by the companion Static HTML Space.
summary.json: aggregate image… See the full description on the dataset page: https://huggingface.co/datasets/zoeloopy/lewm-vjepa21-state-moe-shared-residual-wo-tokencl-analysis.cs2-10k-vjepa2-latents-300
CS2-10k V-JEPA2 latent cache (300 matches)
Frozen facebook/vjepa2-vitl-fpc64-256 embeddings of single-POV Counter-Strike 2
gameplay, from 300 matches of RekaAI/CS2-10k
(mirage + dust2). This is the training substrate for the
scale300 hierarchical world model.
Rebuilding it from source takes ~82 hours of wall-clock (elapsed_min 4942.7),
almost all of it network-bound, which is why it is published here.
Contents
file
shape / rows
notes
latents.npy
[6,942… See the full description on the dataset page: https://huggingface.co/datasets/cbctr/cs2-10k-vjepa2-latents-300.aloha_real_agilex_folding_fabric-vjepa2-pool
aloha_real_agilex_folding_fabric with pooled V-JEPA2.1 targets
This is a LeRobot v2.1 copy of SakikoTogawa/aloha_real_agilex_folding_fabric with observation.concept added.
The 2304-dimensional target follows vjepa2-1-vitb384-spatialpool-t0-3concat: for every timestep and each
camera in observation.images.cam_left_wrist, observation.images.cam_right_wrist, observation.images.cam_high, a 64-frame forward clip is resized to 384x384,
encoded with vjepa2_1_vit_base_384, spatially… See the full description on the dataset page: https://huggingface.co/datasets/JackieMM/aloha_real_agilex_folding_fabric-vjepa2-pool.vjepa2-reacher-world-modelvjepa2-temporal-order-blindspots
V-JEPA2 Temporal Order Blind Spots
This dataset documents blind spots for the pretrained video world model facebook/vjepa2-vith-fpc64-256.
The model produces nearly identical embeddings for videos whose temporal order has been severely corrupted, indicating weak sensitivity to temporal directionality and causal motion structure.
Model Tested
Model name: facebook/vjepa2-vith-fpc64-256Type: Self-supervised video world model (JEPA-style joint embedding predictive… See the full description on the dataset page: https://huggingface.co/datasets/Nuntea/vjepa2-temporal-order-blindspots.bmd-vjepa2-featuressokoban-10k-vjepa2-tokenized-shardscs280-synthetic-circles-v3-embeddings-vjepa21-vitg-384_t16cs280-synthetic-circles-v4-embeddings-vjepa21-vitg-384-t16cs280-synthetic-circles-v2-embeddings-vjepa21-vitg-384vjepa2-bridge-embeddingscs280-synthetic-circles-v2-embeddings-vjepa2