| { | |
| "encoder": "answerdotai/ModernBERT-large", | |
| "head_layers": 2, | |
| "max_len": 512, | |
| "head_max_len": 192, | |
| "max_prefixes": 6, | |
| "act_costs": { | |
| "escalate": 0.5 | |
| }, | |
| "cost_wrong_act": 3.0, | |
| "amp_dtype": "bf16", | |
| "model_name": "rl-agent", | |
| "temperature": [ | |
| 1.2236367464065552, | |
| 1.1360598802566528, | |
| 1.1707336902618408 | |
| ], | |
| "temperature_by_options": {}, | |
| "training": { | |
| "updates": 7313, | |
| "epochs_completed": 1, | |
| "hours": 1.96, | |
| "world_size": 1, | |
| "fine_tuned_from_checkpoint": true | |
| }, | |
| "vision": { | |
| "tower": "google/siglip-base-patch16-224", | |
| "tower_revision": "7fd15f0689c79d79e38b1c2e2e2370a7bf2761ed", | |
| "feature_layer": -2, | |
| "pool_mode": "avg", | |
| "pool_k": 2, | |
| "image_size": 224, | |
| "patch_size": 16, | |
| "vision_width": 768, | |
| "tower_trained": false, | |
| "proj_in_norm": true, | |
| "proj_standardize": true, | |
| "wise_alpha": 0.85, | |
| "temperature": [ | |
| 1.2782474756240845, | |
| 1.2514300346374512, | |
| 1.36097252368927 | |
| ], | |
| "temperature_by_options": { | |
| "choice:11+": 1.337010383605957, | |
| "choice:3-5": 1.306784749031067, | |
| "choice:6-10": 1.1875139474868774, | |
| "noul:2": 1.36097252368927 | |
| }, | |
| "laya_base": "convaiinnovations/laya@55cf4c4ebb4ebe31b2550e8bdf3bd21b99753851", | |
| "laya_version": "0.3.21", | |
| "n_tokens": 49 | |
| } | |
| } |