Upload prism-qwen25-dinosiglip-224px+0_5b+mx-roboverse+n0+b32+x7-stackcube-vq7/config.json with huggingface_hub

prism-qwen25-dinosiglip-224px+0_5b+mx-roboverse+n0+b32+x7-stackcube-vq7/config.json

@@ -0,0 +1,51 @@
+{
+  "data_root_dir": "stackcube_new_render/1.0.0",
+  "expected_world_size": 8,
+  "global_batch_size": 384,
+  "hf_token": ".hf_token",
+  "image_aug": false,
+  "is_resume": true,
+  "per_device_batch_size": 48,
+  "pretrained_checkpoint": "logs/prism-qwen25-dinosiglip-224px+0_5b+mx-roboverse+n0+b32+x7-stackcube-vq7/checkpoints/step-039540-epoch-20-loss=0.2883.pt",
+  "resume_epoch": 20,
+  "resume_step": 39540,
+  "run_id": "prism-qwen25-dinosiglip-224px+0_5b+mx-roboverse+n0+b32+x7-stackcube-vq7",
+  "run_id_note": null,
+  "run_root_dir": "logs",
+  "save_interval": 2500,
+  "seed": 7,
+  "trackers": [
+    "jsonl",
+    "wandb"
+  ],
+  "vla": {
+    "action_tokenizer": "stackcube_vq_extra_action_tokenizer",
+    "base_vlm": "prism-qwen25-extra-dinosiglip-224px+0_5b",
+    "data_mix": "roboverse",
+    "enable_gradient_checkpointing": true,
+    "enable_mixed_precision_training": true,
+    "epochs": 40,
+    "expected_world_size": 8,
+    "freeze_llm_backbone": false,
+    "freeze_vision_backbone": false,
+    "global_batch_size": 384,
+    "image_sequence_len": 1,
+    "learning_rate": 2e-05,
+    "lr_scheduler_type": "constant",
+    "max_grad_norm": 1.0,
+    "max_steps": null,
+    "per_device_batch_size": 48,
+    "reduce_in_full_precision": true,
+    "save_every_n_steps": 25000,
+    "shuffle_buffer_size": 256000,
+    "train_strategy": "fsdp-full-shard",
+    "type": "prism-qwen25-dinosiglip-224px+0_5b+mx-roboverse",
+    "unfreeze_last_llm_layer": false,
+    "use_wrist_image": false,
+    "vla_id": "prism-qwen25-dinosiglip-224px+0_5b+mx-roboverse",
+    "warmup_ratio": 0.0,
+    "weight_decay": 0.0
+  },
+  "wandb_entity": "szang",
+  "wandb_project": "openvla"
+}