| { | |
| "model_config": { | |
| "input_dim": 2048, | |
| "hidden_dim": 256, | |
| "projection_dim": 128 | |
| }, | |
| "encoder": { | |
| "model_id": "Qwen/Qwen3-1.7B", | |
| "revision": "70d244cc86ccca08cf5af4e1e306ecf908b1ad5e", | |
| "recipe": "plain-state-question-v1:last-nonpad:l2:reject-truncation", | |
| "max_length": 256, | |
| "workers": 2, | |
| "inventory_hash": "5be782b4e33e4c9a9b5c70a71f3a2938fa873285c6c8382ed32ddf05b7b99068", | |
| "total_texts": 8010, | |
| "dtype": "float16" | |
| }, | |
| "temperature": 1.0, | |
| "epoch": 2, | |
| "training_method": "supervised", | |
| "data_sha256": "03754f8be44e761d329beb6fa6106c4f6ce8184bcddb6d1768ad054b3aab1dae", | |
| "settings": { | |
| "method": "supervised", | |
| "epochs": 12, | |
| "batch_size": 128, | |
| "lr": 0.0001, | |
| "samples": 8, | |
| "sigma_start": 0.5, | |
| "sigma_end": 0.05, | |
| "seed": 42, | |
| "device": "cuda:0", | |
| "reward": "negative_multiclass_brier", | |
| "policy": "gaussian_latent_softmax_v1" | |
| }, | |
| "format": "tiny-jev-heads-v1", | |
| "source_checkpoint_sha256": "61b7a77120a14479c1a8c3e9d13289f224e1a843a34960a4b65cd932273a910d", | |
| "weights_sha256": "c0fbf1968f333f01ecd4b240a781b69dec9e10432e42ccb12350fcef23c9fb5e", | |
| "trainable_parameters": 1123073, | |
| "release_status": "experimental research preview; updated adapter Hub/GPU check pending", | |
| "license": "apache-2.0", | |
| "maintainer": "sainath", | |
| "hub_repo_id": "sainath/tiny-jev-support-v03-supervised", | |
| "loader_version": "0.4.0", | |
| "interface": "jev-shaped-subset-v1" | |
| } | |