| { | |
| "best_metric": 0.26981785893440247, | |
| "best_model_checkpoint": "miner_id_24/checkpoint-500", | |
| "epoch": 0.09550186228631459, | |
| "eval_steps": 50, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00019100372457262916, | |
| "eval_loss": 2.06730318069458, | |
| "eval_runtime": 63.62, | |
| "eval_samples_per_second": 34.659, | |
| "eval_steps_per_second": 8.677, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0019100372457262916, | |
| "grad_norm": 7.6455230712890625, | |
| "learning_rate": 4.0400000000000006e-05, | |
| "loss": 3.1643, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.003820074491452583, | |
| "grad_norm": 4.942553997039795, | |
| "learning_rate": 8.080000000000001e-05, | |
| "loss": 2.6878, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.005730111737178875, | |
| "grad_norm": 4.34798526763916, | |
| "learning_rate": 0.00012119999999999999, | |
| "loss": 1.7231, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.007640148982905166, | |
| "grad_norm": 5.313558578491211, | |
| "learning_rate": 0.00016160000000000002, | |
| "loss": 1.7659, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.009550186228631458, | |
| "grad_norm": 6.8922014236450195, | |
| "learning_rate": 0.000202, | |
| "loss": 1.3088, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.009550186228631458, | |
| "eval_loss": 0.8562319278717041, | |
| "eval_runtime": 63.8502, | |
| "eval_samples_per_second": 34.534, | |
| "eval_steps_per_second": 8.645, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01146022347435775, | |
| "grad_norm": 3.368533134460449, | |
| "learning_rate": 0.00020175396907624226, | |
| "loss": 1.4345, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.013370260720084041, | |
| "grad_norm": 3.520836591720581, | |
| "learning_rate": 0.0002010170749428986, | |
| "loss": 1.3208, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.015280297965810333, | |
| "grad_norm": 3.200895309448242, | |
| "learning_rate": 0.00019979290767411438, | |
| "loss": 1.2472, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.017190335211536626, | |
| "grad_norm": 3.8880531787872314, | |
| "learning_rate": 0.0001980874312897702, | |
| "loss": 0.9301, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.019100372457262916, | |
| "grad_norm": 5.285026550292969, | |
| "learning_rate": 0.00019590895469937675, | |
| "loss": 1.0703, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.019100372457262916, | |
| "eval_loss": 0.6190745830535889, | |
| "eval_runtime": 63.8537, | |
| "eval_samples_per_second": 34.532, | |
| "eval_steps_per_second": 8.645, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.02101040970298921, | |
| "grad_norm": 2.872575283050537, | |
| "learning_rate": 0.0001932680912219027, | |
| "loss": 1.0179, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.0229204469487155, | |
| "grad_norm": 3.166882038116455, | |
| "learning_rate": 0.00019017770687875164, | |
| "loss": 0.9739, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.024830484194441792, | |
| "grad_norm": 3.3663108348846436, | |
| "learning_rate": 0.000186652857711799, | |
| "loss": 0.8537, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.026740521440168082, | |
| "grad_norm": 2.7930142879486084, | |
| "learning_rate": 0.00018271071643186968, | |
| "loss": 0.7789, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.028650558685894376, | |
| "grad_norm": 4.935205459594727, | |
| "learning_rate": 0.00017837048875501678, | |
| "loss": 1.0222, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.028650558685894376, | |
| "eval_loss": 0.5046119689941406, | |
| "eval_runtime": 63.8937, | |
| "eval_samples_per_second": 34.51, | |
| "eval_steps_per_second": 8.639, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.030560595931620665, | |
| "grad_norm": 2.4444825649261475, | |
| "learning_rate": 0.00017365331983420376, | |
| "loss": 0.9172, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.032470633177346955, | |
| "grad_norm": 3.213865280151367, | |
| "learning_rate": 0.0001685821912422447, | |
| "loss": 0.8119, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.03438067042307325, | |
| "grad_norm": 2.578925371170044, | |
| "learning_rate": 0.00016318180900789148, | |
| "loss": 0.7952, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.03629070766879954, | |
| "grad_norm": 3.1453428268432617, | |
| "learning_rate": 0.00015747848325054544, | |
| "loss": 0.774, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.03820074491452583, | |
| "grad_norm": 6.084349155426025, | |
| "learning_rate": 0.0001515, | |
| "loss": 0.8455, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.03820074491452583, | |
| "eval_loss": 0.4781758189201355, | |
| "eval_runtime": 63.8211, | |
| "eval_samples_per_second": 34.55, | |
| "eval_steps_per_second": 8.649, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04011078216025212, | |
| "grad_norm": 2.6562137603759766, | |
| "learning_rate": 0.00014527548582569683, | |
| "loss": 0.9388, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.04202081940597842, | |
| "grad_norm": 2.2333340644836426, | |
| "learning_rate": 0.00013883526593500714, | |
| "loss": 0.7923, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.04393085665170471, | |
| "grad_norm": 2.876622200012207, | |
| "learning_rate": 0.0001322107164318697, | |
| "loss": 0.8484, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.045840893897431, | |
| "grad_norm": 3.026136636734009, | |
| "learning_rate": 0.00012543411145556643, | |
| "loss": 0.714, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.047750931143157295, | |
| "grad_norm": 4.8779826164245605, | |
| "learning_rate": 0.00011853846594435998, | |
| "loss": 0.6556, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.047750931143157295, | |
| "eval_loss": 0.3865682780742645, | |
| "eval_runtime": 63.8484, | |
| "eval_samples_per_second": 34.535, | |
| "eval_steps_per_second": 8.645, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.049660968388883585, | |
| "grad_norm": 2.4860901832580566, | |
| "learning_rate": 0.00011155737479003301, | |
| "loss": 0.634, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.051571005634609875, | |
| "grad_norm": 1.8885411024093628, | |
| "learning_rate": 0.00010452484916695262, | |
| "loss": 0.7666, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.053481042880336165, | |
| "grad_norm": 3.745715856552124, | |
| "learning_rate": 9.747515083304742e-05, | |
| "loss": 0.742, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.05539108012606246, | |
| "grad_norm": 2.1745243072509766, | |
| "learning_rate": 9.044262520996702e-05, | |
| "loss": 0.6505, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.05730111737178875, | |
| "grad_norm": 3.0038623809814453, | |
| "learning_rate": 8.346153405564004e-05, | |
| "loss": 0.6772, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.05730111737178875, | |
| "eval_loss": 0.34666135907173157, | |
| "eval_runtime": 63.8243, | |
| "eval_samples_per_second": 34.548, | |
| "eval_steps_per_second": 8.649, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.05921115461751504, | |
| "grad_norm": 1.8677786588668823, | |
| "learning_rate": 7.656588854443357e-05, | |
| "loss": 0.5809, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.06112119186324133, | |
| "grad_norm": 1.749708652496338, | |
| "learning_rate": 6.978928356813031e-05, | |
| "loss": 0.5507, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.06303122910896762, | |
| "grad_norm": 2.1635003089904785, | |
| "learning_rate": 6.316473406499288e-05, | |
| "loss": 0.5366, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.06494126635469391, | |
| "grad_norm": 3.309890031814575, | |
| "learning_rate": 5.672451417430317e-05, | |
| "loss": 0.6328, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.06685130360042021, | |
| "grad_norm": 4.125185966491699, | |
| "learning_rate": 5.050000000000002e-05, | |
| "loss": 0.5743, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.06685130360042021, | |
| "eval_loss": 0.3164260983467102, | |
| "eval_runtime": 63.6764, | |
| "eval_samples_per_second": 34.628, | |
| "eval_steps_per_second": 8.669, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.0687613408461465, | |
| "grad_norm": 1.8781318664550781, | |
| "learning_rate": 4.452151674945458e-05, | |
| "loss": 0.6682, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.0706713780918728, | |
| "grad_norm": 1.7442470788955688, | |
| "learning_rate": 3.8818190992108515e-05, | |
| "loss": 0.5477, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.07258141533759908, | |
| "grad_norm": 2.8514626026153564, | |
| "learning_rate": 3.3417808757755355e-05, | |
| "loss": 0.5742, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.07449145258332537, | |
| "grad_norm": 2.2705271244049072, | |
| "learning_rate": 2.8346680165796253e-05, | |
| "loss": 0.6682, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.07640148982905166, | |
| "grad_norm": 4.242289066314697, | |
| "learning_rate": 2.362951124498323e-05, | |
| "loss": 0.7781, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.07640148982905166, | |
| "eval_loss": 0.288888543844223, | |
| "eval_runtime": 63.7944, | |
| "eval_samples_per_second": 34.564, | |
| "eval_steps_per_second": 8.653, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.07831152707477795, | |
| "grad_norm": 2.7676401138305664, | |
| "learning_rate": 1.928928356813032e-05, | |
| "loss": 0.4907, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.08022156432050424, | |
| "grad_norm": 2.335130214691162, | |
| "learning_rate": 1.5347142288200977e-05, | |
| "loss": 0.6022, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.08213160156623055, | |
| "grad_norm": 2.4311466217041016, | |
| "learning_rate": 1.1822293121248375e-05, | |
| "loss": 0.5102, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.08404163881195684, | |
| "grad_norm": 2.3990116119384766, | |
| "learning_rate": 8.731908778097302e-06, | |
| "loss": 0.6047, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.08595167605768313, | |
| "grad_norm": 4.477560043334961, | |
| "learning_rate": 6.09104530062326e-06, | |
| "loss": 0.7265, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.08595167605768313, | |
| "eval_loss": 0.27164798974990845, | |
| "eval_runtime": 63.8913, | |
| "eval_samples_per_second": 34.512, | |
| "eval_steps_per_second": 8.64, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.08786171330340942, | |
| "grad_norm": 2.161862373352051, | |
| "learning_rate": 3.912568710229791e-06, | |
| "loss": 0.5218, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.0897717505491357, | |
| "grad_norm": 1.7147796154022217, | |
| "learning_rate": 2.2070923258856255e-06, | |
| "loss": 0.4866, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.091681787794862, | |
| "grad_norm": 2.2955782413482666, | |
| "learning_rate": 9.829250571013935e-07, | |
| "loss": 0.516, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.09359182504058829, | |
| "grad_norm": 3.3165369033813477, | |
| "learning_rate": 2.4603092375775605e-07, | |
| "loss": 0.4647, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.09550186228631459, | |
| "grad_norm": 2.52410888671875, | |
| "learning_rate": 0.0, | |
| "loss": 0.5188, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09550186228631459, | |
| "eval_loss": 0.26981785893440247, | |
| "eval_runtime": 64.0421, | |
| "eval_samples_per_second": 34.43, | |
| "eval_steps_per_second": 8.619, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.300574941413376e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |