| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9838187702265372, | |
| "eval_steps": 500, | |
| "global_step": 38, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.025889967637540454, | |
| "grad_norm": 14.94955365537486, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6643, | |
| "mean_token_accuracy": 0.8553724065423012, | |
| "num_tokens": 65536.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05177993527508091, | |
| "grad_norm": 13.315623180611107, | |
| "learning_rate": 1.9473684210526318e-05, | |
| "loss": 0.452, | |
| "mean_token_accuracy": 0.8764078170061111, | |
| "num_tokens": 130612.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.07766990291262135, | |
| "grad_norm": 3.2684775267493738, | |
| "learning_rate": 1.894736842105263e-05, | |
| "loss": 0.3998, | |
| "mean_token_accuracy": 0.8837301507592201, | |
| "num_tokens": 196148.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.10355987055016182, | |
| "grad_norm": 3.648625597879628, | |
| "learning_rate": 1.8421052631578947e-05, | |
| "loss": 0.4149, | |
| "mean_token_accuracy": 0.8796855807304382, | |
| "num_tokens": 261684.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.12944983818770225, | |
| "grad_norm": 2.1735962344366744, | |
| "learning_rate": 1.7894736842105264e-05, | |
| "loss": 0.4029, | |
| "mean_token_accuracy": 0.8861111029982567, | |
| "num_tokens": 327220.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.1553398058252427, | |
| "grad_norm": 1.2399813671584698, | |
| "learning_rate": 1.736842105263158e-05, | |
| "loss": 0.3874, | |
| "mean_token_accuracy": 0.8873015865683556, | |
| "num_tokens": 392756.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.18122977346278318, | |
| "grad_norm": 1.2840669625935126, | |
| "learning_rate": 1.6842105263157896e-05, | |
| "loss": 0.3925, | |
| "mean_token_accuracy": 0.8855463936924934, | |
| "num_tokens": 458292.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.20711974110032363, | |
| "grad_norm": 0.9666205702654993, | |
| "learning_rate": 1.6315789473684213e-05, | |
| "loss": 0.4035, | |
| "mean_token_accuracy": 0.8799450471997261, | |
| "num_tokens": 523828.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.23300970873786409, | |
| "grad_norm": 1.0109942942154606, | |
| "learning_rate": 1.578947368421053e-05, | |
| "loss": 0.434, | |
| "mean_token_accuracy": 0.8735195249319077, | |
| "num_tokens": 589364.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2588996763754045, | |
| "grad_norm": 1.0413113875886446, | |
| "learning_rate": 1.5263157894736846e-05, | |
| "loss": 0.4191, | |
| "mean_token_accuracy": 0.8769078031182289, | |
| "num_tokens": 654900.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.284789644012945, | |
| "grad_norm": 1.0902203237712937, | |
| "learning_rate": 1.4736842105263159e-05, | |
| "loss": 0.4493, | |
| "mean_token_accuracy": 0.8699328452348709, | |
| "num_tokens": 720436.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.3106796116504854, | |
| "grad_norm": 0.8849467554405712, | |
| "learning_rate": 1.4210526315789475e-05, | |
| "loss": 0.3931, | |
| "mean_token_accuracy": 0.8852578699588776, | |
| "num_tokens": 785579.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.3365695792880259, | |
| "grad_norm": 1.035926324924405, | |
| "learning_rate": 1.3684210526315791e-05, | |
| "loss": 0.39, | |
| "mean_token_accuracy": 0.8831959590315819, | |
| "num_tokens": 851115.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.36245954692556637, | |
| "grad_norm": 1.0026322980151723, | |
| "learning_rate": 1.3157894736842108e-05, | |
| "loss": 0.4184, | |
| "mean_token_accuracy": 0.8782203868031502, | |
| "num_tokens": 916651.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.3883495145631068, | |
| "grad_norm": 1.0102817872020193, | |
| "learning_rate": 1.263157894736842e-05, | |
| "loss": 0.3938, | |
| "mean_token_accuracy": 0.8836233168840408, | |
| "num_tokens": 982187.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.41423948220064727, | |
| "grad_norm": 0.9944421404980388, | |
| "learning_rate": 1.2105263157894737e-05, | |
| "loss": 0.4384, | |
| "mean_token_accuracy": 0.8699938952922821, | |
| "num_tokens": 1047723.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.4401294498381877, | |
| "grad_norm": 0.8998162798325435, | |
| "learning_rate": 1.1578947368421053e-05, | |
| "loss": 0.3713, | |
| "mean_token_accuracy": 0.8903498351573944, | |
| "num_tokens": 1112544.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.46601941747572817, | |
| "grad_norm": 0.9087432992292579, | |
| "learning_rate": 1.105263157894737e-05, | |
| "loss": 0.4036, | |
| "mean_token_accuracy": 0.8820064589381218, | |
| "num_tokens": 1174265.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.4919093851132686, | |
| "grad_norm": 0.8697071517631878, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 0.3931, | |
| "mean_token_accuracy": 0.8835927844047546, | |
| "num_tokens": 1239801.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.517799352750809, | |
| "grad_norm": 0.8774544351292555, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4133, | |
| "mean_token_accuracy": 0.8790166154503822, | |
| "num_tokens": 1303377.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5436893203883495, | |
| "grad_norm": 0.994546197670681, | |
| "learning_rate": 9.473684210526315e-06, | |
| "loss": 0.4375, | |
| "mean_token_accuracy": 0.8726342990994453, | |
| "num_tokens": 1368913.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.56957928802589, | |
| "grad_norm": 0.9232673730081883, | |
| "learning_rate": 8.947368421052632e-06, | |
| "loss": 0.4048, | |
| "mean_token_accuracy": 0.8793192803859711, | |
| "num_tokens": 1434449.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.5954692556634305, | |
| "grad_norm": 1.0209861898624548, | |
| "learning_rate": 8.421052631578948e-06, | |
| "loss": 0.397, | |
| "mean_token_accuracy": 0.8836233168840408, | |
| "num_tokens": 1499985.0, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.6213592233009708, | |
| "grad_norm": 0.9046118944574179, | |
| "learning_rate": 7.894736842105265e-06, | |
| "loss": 0.4138, | |
| "mean_token_accuracy": 0.8774877861142159, | |
| "num_tokens": 1565521.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.6472491909385113, | |
| "grad_norm": 0.9351060177767945, | |
| "learning_rate": 7.368421052631579e-06, | |
| "loss": 0.3922, | |
| "mean_token_accuracy": 0.8843711838126183, | |
| "num_tokens": 1631057.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.6731391585760518, | |
| "grad_norm": 0.9167336580555596, | |
| "learning_rate": 6.842105263157896e-06, | |
| "loss": 0.377, | |
| "mean_token_accuracy": 0.8888125643134117, | |
| "num_tokens": 1696593.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.6990291262135923, | |
| "grad_norm": 0.8827610291370933, | |
| "learning_rate": 6.31578947368421e-06, | |
| "loss": 0.4382, | |
| "mean_token_accuracy": 0.8711233139038086, | |
| "num_tokens": 1762129.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7249190938511327, | |
| "grad_norm": 0.8216978597200468, | |
| "learning_rate": 5.789473684210527e-06, | |
| "loss": 0.3978, | |
| "mean_token_accuracy": 0.8840201497077942, | |
| "num_tokens": 1827665.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.7508090614886731, | |
| "grad_norm": 0.7861852214672553, | |
| "learning_rate": 5.263157894736842e-06, | |
| "loss": 0.361, | |
| "mean_token_accuracy": 0.8924908339977264, | |
| "num_tokens": 1893201.0, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.7766990291262136, | |
| "grad_norm": 0.9358327841683868, | |
| "learning_rate": 4.736842105263158e-06, | |
| "loss": 0.3654, | |
| "mean_token_accuracy": 0.89267398416996, | |
| "num_tokens": 1958737.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8025889967637541, | |
| "grad_norm": 0.7893903886446123, | |
| "learning_rate": 4.210526315789474e-06, | |
| "loss": 0.3375, | |
| "mean_token_accuracy": 0.8997863158583641, | |
| "num_tokens": 2024273.0, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.8284789644012945, | |
| "grad_norm": 0.8243141967647842, | |
| "learning_rate": 3.6842105263157896e-06, | |
| "loss": 0.3728, | |
| "mean_token_accuracy": 0.8913461416959763, | |
| "num_tokens": 2089809.0, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.8543689320388349, | |
| "grad_norm": 0.8246709048525032, | |
| "learning_rate": 3.157894736842105e-06, | |
| "loss": 0.3643, | |
| "mean_token_accuracy": 0.8915140256285667, | |
| "num_tokens": 2155345.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.8802588996763754, | |
| "grad_norm": 0.8260415283504307, | |
| "learning_rate": 2.631578947368421e-06, | |
| "loss": 0.3828, | |
| "mean_token_accuracy": 0.8868742287158966, | |
| "num_tokens": 2220881.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.9061488673139159, | |
| "grad_norm": 0.7747154448934308, | |
| "learning_rate": 2.105263157894737e-06, | |
| "loss": 0.3455, | |
| "mean_token_accuracy": 0.897390104830265, | |
| "num_tokens": 2286417.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.9320388349514563, | |
| "grad_norm": 0.8212045075297872, | |
| "learning_rate": 1.5789473684210526e-06, | |
| "loss": 0.4002, | |
| "mean_token_accuracy": 0.8831327557563782, | |
| "num_tokens": 2348817.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.9579288025889967, | |
| "grad_norm": 0.8299799490223582, | |
| "learning_rate": 1.0526315789473685e-06, | |
| "loss": 0.442, | |
| "mean_token_accuracy": 0.8720848485827446, | |
| "num_tokens": 2414353.0, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.9838187702265372, | |
| "grad_norm": 0.8097572023142755, | |
| "learning_rate": 5.263157894736843e-07, | |
| "loss": 0.3812, | |
| "mean_token_accuracy": 0.8855026215314865, | |
| "num_tokens": 2478526.0, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.9838187702265372, | |
| "step": 38, | |
| "total_flos": 4940351791104.0, | |
| "train_loss": 0.40646139257832575, | |
| "train_runtime": 867.7069, | |
| "train_samples_per_second": 0.711, | |
| "train_steps_per_second": 0.044 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 38, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4940351791104.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |