| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.24040267447975358, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.002404026744797536, | |
| "eval_loss": 3.228058099746704, | |
| "eval_runtime": 156.333, | |
| "eval_samples_per_second": 8.968, | |
| "eval_steps_per_second": 4.484, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007212080234392608, | |
| "grad_norm": 1.9650676250457764, | |
| "learning_rate": 3e-05, | |
| "loss": 2.349, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.014424160468785216, | |
| "grad_norm": 1.4064522981643677, | |
| "learning_rate": 6e-05, | |
| "loss": 2.2216, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.021636240703177823, | |
| "grad_norm": 1.3426527976989746, | |
| "learning_rate": 9e-05, | |
| "loss": 1.9405, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.021636240703177823, | |
| "eval_loss": 1.9358376264572144, | |
| "eval_runtime": 156.5558, | |
| "eval_samples_per_second": 8.955, | |
| "eval_steps_per_second": 4.478, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.02884832093757043, | |
| "grad_norm": 1.1980935335159302, | |
| "learning_rate": 0.00012, | |
| "loss": 1.7996, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03606040117196304, | |
| "grad_norm": 0.9815410375595093, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.6674, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.043272481406355645, | |
| "grad_norm": 1.1842740774154663, | |
| "learning_rate": 0.00018, | |
| "loss": 1.5861, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.043272481406355645, | |
| "eval_loss": 1.4581845998764038, | |
| "eval_runtime": 156.5288, | |
| "eval_samples_per_second": 8.957, | |
| "eval_steps_per_second": 4.478, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.05048456164074825, | |
| "grad_norm": 0.9527170062065125, | |
| "learning_rate": 0.0001999229036240723, | |
| "loss": 1.5801, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.05769664187514086, | |
| "grad_norm": 0.9741354584693909, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 1.6089, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06490872210953347, | |
| "grad_norm": 0.8610622882843018, | |
| "learning_rate": 0.00019624552364536473, | |
| "loss": 1.5059, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.06490872210953347, | |
| "eval_loss": 1.3450061082839966, | |
| "eval_runtime": 156.5047, | |
| "eval_samples_per_second": 8.958, | |
| "eval_steps_per_second": 4.479, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.07212080234392608, | |
| "grad_norm": 0.988355815410614, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 1.4597, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07933288257831868, | |
| "grad_norm": 0.9052969813346863, | |
| "learning_rate": 0.00018724960070727972, | |
| "loss": 1.3148, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.08654496281271129, | |
| "grad_norm": 0.9871351718902588, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 1.4494, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.08654496281271129, | |
| "eval_loss": 1.2605727910995483, | |
| "eval_runtime": 156.5558, | |
| "eval_samples_per_second": 8.955, | |
| "eval_steps_per_second": 4.478, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0937570430471039, | |
| "grad_norm": 1.216485619544983, | |
| "learning_rate": 0.00017343225094356855, | |
| "loss": 1.3575, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.1009691232814965, | |
| "grad_norm": 1.0345182418823242, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 1.316, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.10818120351588911, | |
| "grad_norm": 1.2272484302520752, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 1.3834, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.10818120351588911, | |
| "eval_loss": 1.199688196182251, | |
| "eval_runtime": 156.5688, | |
| "eval_samples_per_second": 8.955, | |
| "eval_steps_per_second": 4.477, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.11539328375028172, | |
| "grad_norm": 0.8153639435768127, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 1.3282, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.12260536398467432, | |
| "grad_norm": 0.9661998748779297, | |
| "learning_rate": 0.0001346117057077493, | |
| "loss": 1.2335, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.12981744421906694, | |
| "grad_norm": 0.9782885313034058, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 1.1531, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.12981744421906694, | |
| "eval_loss": 1.1464760303497314, | |
| "eval_runtime": 156.5981, | |
| "eval_samples_per_second": 8.953, | |
| "eval_steps_per_second": 4.476, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.13702952445345953, | |
| "grad_norm": 1.0093810558319092, | |
| "learning_rate": 0.00011175373974578378, | |
| "loss": 1.1367, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.14424160468785216, | |
| "grad_norm": 0.9035248160362244, | |
| "learning_rate": 0.0001, | |
| "loss": 1.2007, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.15145368492224476, | |
| "grad_norm": 1.1476191282272339, | |
| "learning_rate": 8.824626025421626e-05, | |
| "loss": 1.1702, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.15145368492224476, | |
| "eval_loss": 1.1219838857650757, | |
| "eval_runtime": 156.5351, | |
| "eval_samples_per_second": 8.956, | |
| "eval_steps_per_second": 4.478, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.15866576515663736, | |
| "grad_norm": 1.004503607749939, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 1.1231, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.16587784539102998, | |
| "grad_norm": 0.8763646483421326, | |
| "learning_rate": 6.538829429225069e-05, | |
| "loss": 1.1959, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.17308992562542258, | |
| "grad_norm": 1.0786497592926025, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 1.2195, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.17308992562542258, | |
| "eval_loss": 1.0894047021865845, | |
| "eval_runtime": 156.5655, | |
| "eval_samples_per_second": 8.955, | |
| "eval_steps_per_second": 4.477, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.18030200585981518, | |
| "grad_norm": 1.005875587463379, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 1.0467, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1875140860942078, | |
| "grad_norm": 0.9949033260345459, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 1.3012, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.1947261663286004, | |
| "grad_norm": 0.9470193982124329, | |
| "learning_rate": 2.6567749056431467e-05, | |
| "loss": 1.237, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.1947261663286004, | |
| "eval_loss": 1.070177674293518, | |
| "eval_runtime": 156.4864, | |
| "eval_samples_per_second": 8.959, | |
| "eval_steps_per_second": 4.48, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.201938246562993, | |
| "grad_norm": 0.9711623787879944, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.9719, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.20915032679738563, | |
| "grad_norm": 1.0642070770263672, | |
| "learning_rate": 1.2750399292720283e-05, | |
| "loss": 1.2356, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.21636240703177823, | |
| "grad_norm": 0.8378780484199524, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 1.1906, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.21636240703177823, | |
| "eval_loss": 1.0622040033340454, | |
| "eval_runtime": 156.5263, | |
| "eval_samples_per_second": 8.957, | |
| "eval_steps_per_second": 4.478, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.22357448726617082, | |
| "grad_norm": 0.9879840016365051, | |
| "learning_rate": 3.7544763546352834e-06, | |
| "loss": 1.1005, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.23078656750056345, | |
| "grad_norm": 0.884789764881134, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 1.0352, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.23799864773495605, | |
| "grad_norm": 1.1920340061187744, | |
| "learning_rate": 7.709637592770991e-08, | |
| "loss": 1.117, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.23799864773495605, | |
| "eval_loss": 1.0599479675292969, | |
| "eval_runtime": 156.5531, | |
| "eval_samples_per_second": 8.955, | |
| "eval_steps_per_second": 4.478, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.969948643447276e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |