| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 340, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14705882352941177, | |
| "grad_norm": 2.458341360092163, | |
| "learning_rate": 0.00011764705882352942, | |
| "loss": 0.8805, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.29411764705882354, | |
| "grad_norm": 1.3460900783538818, | |
| "learning_rate": 0.00019995743267092514, | |
| "loss": 0.4012, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4411764705882353, | |
| "grad_norm": 1.291209101676941, | |
| "learning_rate": 0.0001992016878768271, | |
| "loss": 0.3084, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 1.0544739961624146, | |
| "learning_rate": 0.00019750822700209488, | |
| "loss": 0.2579, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.7352941176470589, | |
| "grad_norm": 1.0863096714019775, | |
| "learning_rate": 0.00019489305769264812, | |
| "loss": 0.241, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8823529411764706, | |
| "grad_norm": 0.8423298001289368, | |
| "learning_rate": 0.00019138090015605118, | |
| "loss": 0.2391, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.0294117647058822, | |
| "grad_norm": 0.7478500008583069, | |
| "learning_rate": 0.0001870049534907294, | |
| "loss": 0.1854, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 0.7334464192390442, | |
| "learning_rate": 0.00018180658186743332, | |
| "loss": 0.1821, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.3235294117647058, | |
| "grad_norm": 1.0398321151733398, | |
| "learning_rate": 0.00017583492352935814, | |
| "loss": 0.164, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.4705882352941178, | |
| "grad_norm": 0.9809992909431458, | |
| "learning_rate": 0.00016914642630689388, | |
| "loss": 0.1647, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.6176470588235294, | |
| "grad_norm": 0.5706636905670166, | |
| "learning_rate": 0.00016180431403760837, | |
| "loss": 0.1683, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 0.7732464671134949, | |
| "learning_rate": 0.00015387798893519455, | |
| "loss": 0.1546, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.9117647058823528, | |
| "grad_norm": 0.7296786308288574, | |
| "learning_rate": 0.00014544237555656216, | |
| "loss": 0.1286, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.0588235294117645, | |
| "grad_norm": 0.6389240622520447, | |
| "learning_rate": 0.00013657721256830506, | |
| "loss": 0.1227, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.2058823529411766, | |
| "grad_norm": 0.46665242314338684, | |
| "learning_rate": 0.0001273662990072083, | |
| "loss": 0.1368, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 0.6480994820594788, | |
| "learning_rate": 0.00011789670215960975, | |
| "loss": 0.1192, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.4292305111885071, | |
| "learning_rate": 0.00010825793454723325, | |
| "loss": 0.1051, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.6470588235294117, | |
| "grad_norm": 0.6904204487800598, | |
| "learning_rate": 9.854110779913537e-05, | |
| "loss": 0.1065, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.7941176470588234, | |
| "grad_norm": 0.6822876334190369, | |
| "learning_rate": 8.883807140789478e-05, | |
| "loss": 0.1107, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "grad_norm": 0.5520216822624207, | |
| "learning_rate": 7.924054451105614e-05, | |
| "loss": 0.1155, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.088235294117647, | |
| "grad_norm": 0.4506421685218811, | |
| "learning_rate": 6.983924890477138e-05, | |
| "loss": 0.0978, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.235294117647059, | |
| "grad_norm": 0.538611650466919, | |
| "learning_rate": 6.072305148493195e-05, | |
| "loss": 0.0987, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.3823529411764706, | |
| "grad_norm": 0.3745705485343933, | |
| "learning_rate": 5.197812422197286e-05, | |
| "loss": 0.0956, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 0.458775132894516, | |
| "learning_rate": 4.368712960978864e-05, | |
| "loss": 0.0928, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.6764705882352944, | |
| "grad_norm": 0.31457090377807617, | |
| "learning_rate": 3.59284392884057e-05, | |
| "loss": 0.0906, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.8235294117647056, | |
| "grad_norm": 0.48355451226234436, | |
| "learning_rate": 2.8775393226475224e-05, | |
| "loss": 0.0907, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.9705882352941178, | |
| "grad_norm": 0.32496538758277893, | |
| "learning_rate": 2.229560646625448e-05, | |
| "loss": 0.0813, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.117647058823529, | |
| "grad_norm": 0.38124969601631165, | |
| "learning_rate": 1.6550329984155112e-05, | |
| "loss": 0.0902, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.264705882352941, | |
| "grad_norm": 0.3412385880947113, | |
| "learning_rate": 1.1593871708401526e-05, | |
| "loss": 0.0727, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.411764705882353, | |
| "grad_norm": 0.36698442697525024, | |
| "learning_rate": 7.473083166700945e-06, | |
| "loss": 0.0796, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.5588235294117645, | |
| "grad_norm": 0.43249696493148804, | |
| "learning_rate": 4.226916616450916e-06, | |
| "loss": 0.0807, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 0.40810427069664, | |
| "learning_rate": 1.8860568437648052e-06, | |
| "loss": 0.0825, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.852941176470588, | |
| "grad_norm": 0.3101021945476532, | |
| "learning_rate": 4.726311117803084e-07, | |
| "loss": 0.082, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.40530508756637573, | |
| "learning_rate": 0.0, | |
| "loss": 0.0693, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 340, | |
| "total_flos": 1.1211231724308e+16, | |
| "train_loss": 0.161671401998576, | |
| "train_runtime": 161.3443, | |
| "train_samples_per_second": 33.717, | |
| "train_steps_per_second": 2.107 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 340, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1211231724308e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |