| { | |
| "best_metric": 1.5939216613769531, | |
| "best_model_checkpoint": "opus-mt-zh-en-finetuned-zhen-checkpoints/checkpoint-25985", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 25985, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 2.777989149093628, | |
| "learning_rate": 1.9961593226861654e-05, | |
| "loss": 0.608, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 2.571350336074829, | |
| "learning_rate": 1.9923109486242066e-05, | |
| "loss": 0.6031, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 6.669718265533447, | |
| "learning_rate": 1.9884625745622478e-05, | |
| "loss": 0.5823, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 3.766803503036499, | |
| "learning_rate": 1.984614200500289e-05, | |
| "loss": 0.625, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 2.617687225341797, | |
| "learning_rate": 1.9807735231864538e-05, | |
| "loss": 0.6324, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 4.26535177230835, | |
| "learning_rate": 1.976932845872619e-05, | |
| "loss": 0.6184, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 5.035365581512451, | |
| "learning_rate": 1.9730844718106602e-05, | |
| "loss": 0.6277, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 6.626140594482422, | |
| "learning_rate": 1.9692360977487014e-05, | |
| "loss": 0.6557, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 3.718008279800415, | |
| "learning_rate": 1.9653877236867426e-05, | |
| "loss": 0.6623, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 4.26861047744751, | |
| "learning_rate": 1.9615470463729075e-05, | |
| "loss": 0.6719, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 5.266055107116699, | |
| "learning_rate": 1.9576986723109487e-05, | |
| "loss": 0.636, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 6.086512565612793, | |
| "learning_rate": 1.95385029824899e-05, | |
| "loss": 0.6585, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 4.0940327644348145, | |
| "learning_rate": 1.950001924187031e-05, | |
| "loss": 0.6341, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 7.52503776550293, | |
| "learning_rate": 1.9461535501250722e-05, | |
| "loss": 0.6589, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 3.5400032997131348, | |
| "learning_rate": 1.9423051760631134e-05, | |
| "loss": 0.6627, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 3.525592565536499, | |
| "learning_rate": 1.9384568020011546e-05, | |
| "loss": 0.639, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 3.1940386295318604, | |
| "learning_rate": 1.9346084279391958e-05, | |
| "loss": 0.6433, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 5.362303733825684, | |
| "learning_rate": 1.930767750625361e-05, | |
| "loss": 0.6714, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 4.885318279266357, | |
| "learning_rate": 1.9269193765634022e-05, | |
| "loss": 0.6334, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 7.104074001312256, | |
| "learning_rate": 1.9230710025014434e-05, | |
| "loss": 0.6496, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 5.086849689483643, | |
| "learning_rate": 1.9192226284394843e-05, | |
| "loss": 0.6647, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 10.493572235107422, | |
| "learning_rate": 1.9153819511256495e-05, | |
| "loss": 0.6609, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 10.206048011779785, | |
| "learning_rate": 1.9115335770636907e-05, | |
| "loss": 0.6806, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 9.489506721496582, | |
| "learning_rate": 1.907685203001732e-05, | |
| "loss": 0.641, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.525002479553223, | |
| "learning_rate": 1.903844525687897e-05, | |
| "loss": 0.6863, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 7.507730007171631, | |
| "learning_rate": 1.9000038483740623e-05, | |
| "loss": 0.6829, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 4.827548980712891, | |
| "learning_rate": 1.8961554743121035e-05, | |
| "loss": 0.6597, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 6.409604072570801, | |
| "learning_rate": 1.8923071002501447e-05, | |
| "loss": 0.6445, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 9.57291316986084, | |
| "learning_rate": 1.8884664229363095e-05, | |
| "loss": 0.6882, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 9.129105567932129, | |
| "learning_rate": 1.8846180488743507e-05, | |
| "loss": 0.6719, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 8.071757316589355, | |
| "learning_rate": 1.880769674812392e-05, | |
| "loss": 0.6668, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 3.636784791946411, | |
| "learning_rate": 1.876921300750433e-05, | |
| "loss": 0.6773, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 6.6433515548706055, | |
| "learning_rate": 1.8730729266884743e-05, | |
| "loss": 0.6776, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 3.0176546573638916, | |
| "learning_rate": 1.8692245526265155e-05, | |
| "loss": 0.6587, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 7.913615703582764, | |
| "learning_rate": 1.8653761785645567e-05, | |
| "loss": 0.6665, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 5.335203647613525, | |
| "learning_rate": 1.861527804502598e-05, | |
| "loss": 0.6684, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 5.51749324798584, | |
| "learning_rate": 1.857679430440639e-05, | |
| "loss": 0.6577, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 4.254025936126709, | |
| "learning_rate": 1.8538310563786802e-05, | |
| "loss": 0.6776, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 3.3869874477386475, | |
| "learning_rate": 1.8499826823167214e-05, | |
| "loss": 0.6351, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 12.704727172851562, | |
| "learning_rate": 1.8461343082547626e-05, | |
| "loss": 0.6904, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 3.478996515274048, | |
| "learning_rate": 1.8422859341928038e-05, | |
| "loss": 0.6829, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 6.513260841369629, | |
| "learning_rate": 1.838437560130845e-05, | |
| "loss": 0.6615, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 5.242546081542969, | |
| "learning_rate": 1.8345891860688862e-05, | |
| "loss": 0.6859, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 6.638239860534668, | |
| "learning_rate": 1.8307408120069274e-05, | |
| "loss": 0.6485, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 8.32187557220459, | |
| "learning_rate": 1.8269001346930923e-05, | |
| "loss": 0.6567, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 3.9929823875427246, | |
| "learning_rate": 1.8230517606311334e-05, | |
| "loss": 0.6996, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 5.426717281341553, | |
| "learning_rate": 1.8192110833172987e-05, | |
| "loss": 0.6529, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 7.328232288360596, | |
| "learning_rate": 1.81536270925534e-05, | |
| "loss": 0.6711, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 6.648090362548828, | |
| "learning_rate": 1.811514335193381e-05, | |
| "loss": 0.7026, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 4.047011375427246, | |
| "learning_rate": 1.8076659611314222e-05, | |
| "loss": 0.6867, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 10.354930877685547, | |
| "learning_rate": 1.803825283817587e-05, | |
| "loss": 0.6718, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.5939216613769531, | |
| "eval_runtime": 11.3331, | |
| "eval_samples_per_second": 351.271, | |
| "eval_steps_per_second": 87.884, | |
| "step": 25985 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 259850, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1029890059665408.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |