| { |
| "architectures": [ |
| "PositionEncoder" |
| ], |
| "model_type": "rukh-position-encoder", |
| "library_name": "rukh", |
| "rukh_version": "0.0.1", |
| "stage": "encoder-mmm-v4", |
| "step": 16000, |
| "params": 38971392, |
| "tokenizer": "moves", |
| "vocab_hash": "527c5dda224cab570cb84da8f7dcda0f43fe53edaf86822f899568f5dd824b46", |
| "data_manifest_sha": "aa0809618dbc372d3cc192fef9dd110dca1ce4aae22dca13ea648c02d78c10b4", |
| "git_sha": "db5d48e5db36a0d24e00cbe21375167b8f9f9c3b", |
| "heads": [ |
| "value", |
| "blunder", |
| "result" |
| ], |
| "pooling": "mean", |
| "pretrained_from": null, |
| "input": "moves", |
| "vocab_size": 2030, |
| "square_vocab": 47, |
| "n_layer": 12, |
| "n_head": 8, |
| "d_model": 512, |
| "d_ff": null, |
| "block": 200, |
| "dropout": 0.1, |
| "pos": "learned", |
| "tie_embeddings": true |
| } |
|
|
