| { | |
| "best_global_step": 340, | |
| "best_metric": 0.7385169267654419, | |
| "best_model_checkpoint": "./mangoes/checkpoint-340", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 340, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.058823529411764705, | |
| "grad_norm": 1.9217854738235474, | |
| "learning_rate": 1.9470588235294118e-05, | |
| "loss": 2.044, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11764705882352941, | |
| "grad_norm": 3.201918363571167, | |
| "learning_rate": 1.888235294117647e-05, | |
| "loss": 2.0161, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.17647058823529413, | |
| "grad_norm": 3.1314334869384766, | |
| "learning_rate": 1.8294117647058824e-05, | |
| "loss": 1.9021, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 2.9427032470703125, | |
| "learning_rate": 1.7705882352941177e-05, | |
| "loss": 1.8208, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.29411764705882354, | |
| "grad_norm": 3.096381425857544, | |
| "learning_rate": 1.711764705882353e-05, | |
| "loss": 1.7553, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.35294117647058826, | |
| "grad_norm": 3.1433303356170654, | |
| "learning_rate": 1.6529411764705883e-05, | |
| "loss": 1.7674, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.4117647058823529, | |
| "grad_norm": 3.0195441246032715, | |
| "learning_rate": 1.594117647058824e-05, | |
| "loss": 1.6242, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 3.748368740081787, | |
| "learning_rate": 1.535294117647059e-05, | |
| "loss": 1.5804, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5294117647058824, | |
| "grad_norm": 2.8905935287475586, | |
| "learning_rate": 1.4764705882352944e-05, | |
| "loss": 1.5126, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 3.1870696544647217, | |
| "learning_rate": 1.4176470588235297e-05, | |
| "loss": 1.3781, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6470588235294118, | |
| "grad_norm": 3.208005428314209, | |
| "learning_rate": 1.3588235294117648e-05, | |
| "loss": 1.4037, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 4.6087236404418945, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 1.2771, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.7647058823529411, | |
| "grad_norm": 3.6908063888549805, | |
| "learning_rate": 1.2411764705882354e-05, | |
| "loss": 1.2711, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.8235294117647058, | |
| "grad_norm": 3.6166765689849854, | |
| "learning_rate": 1.1823529411764707e-05, | |
| "loss": 1.192, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.8823529411764706, | |
| "grad_norm": 3.6934988498687744, | |
| "learning_rate": 1.123529411764706e-05, | |
| "loss": 1.1566, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 3.789727210998535, | |
| "learning_rate": 1.0647058823529413e-05, | |
| "loss": 1.1063, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 3.842630386352539, | |
| "learning_rate": 1.0058823529411766e-05, | |
| "loss": 1.0281, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9583333333333334, | |
| "eval_loss": 1.0490069389343262, | |
| "eval_runtime": 6.306, | |
| "eval_samples_per_second": 38.059, | |
| "eval_steps_per_second": 4.757, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.0588235294117647, | |
| "grad_norm": 3.8409788608551025, | |
| "learning_rate": 9.470588235294119e-06, | |
| "loss": 1.0169, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.1176470588235294, | |
| "grad_norm": 2.461111068725586, | |
| "learning_rate": 8.88235294117647e-06, | |
| "loss": 0.9592, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 3.7472541332244873, | |
| "learning_rate": 8.294117647058825e-06, | |
| "loss": 1.014, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.2352941176470589, | |
| "grad_norm": 4.744520664215088, | |
| "learning_rate": 7.705882352941178e-06, | |
| "loss": 0.9485, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.2941176470588236, | |
| "grad_norm": 2.4809184074401855, | |
| "learning_rate": 7.11764705882353e-06, | |
| "loss": 0.956, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.3529411764705883, | |
| "grad_norm": 6.709966659545898, | |
| "learning_rate": 6.529411764705883e-06, | |
| "loss": 0.9707, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 4.961846828460693, | |
| "learning_rate": 5.941176470588236e-06, | |
| "loss": 0.8539, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.4705882352941178, | |
| "grad_norm": 5.209068298339844, | |
| "learning_rate": 5.352941176470589e-06, | |
| "loss": 0.8284, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.5294117647058822, | |
| "grad_norm": 3.826070547103882, | |
| "learning_rate": 4.764705882352941e-06, | |
| "loss": 0.8226, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.5882352941176472, | |
| "grad_norm": 2.8872721195220947, | |
| "learning_rate": 4.176470588235295e-06, | |
| "loss": 0.7727, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 3.0581214427948, | |
| "learning_rate": 3.5882352941176475e-06, | |
| "loss": 0.7841, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.7058823529411766, | |
| "grad_norm": 4.626227855682373, | |
| "learning_rate": 3e-06, | |
| "loss": 0.7934, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 2.622793436050415, | |
| "learning_rate": 2.411764705882353e-06, | |
| "loss": 0.7713, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.8235294117647058, | |
| "grad_norm": 2.549530267715454, | |
| "learning_rate": 1.8235294117647058e-06, | |
| "loss": 0.7459, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 3.626901149749756, | |
| "learning_rate": 1.235294117647059e-06, | |
| "loss": 0.8056, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.9411764705882353, | |
| "grad_norm": 2.356318950653076, | |
| "learning_rate": 6.470588235294118e-07, | |
| "loss": 0.7665, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.176856517791748, | |
| "learning_rate": 5.882352941176471e-08, | |
| "loss": 0.7454, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9791666666666666, | |
| "eval_loss": 0.7385169267654419, | |
| "eval_runtime": 3.873, | |
| "eval_samples_per_second": 61.967, | |
| "eval_steps_per_second": 7.746, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 340, | |
| "total_flos": 2.1078954658234368e+17, | |
| "train_loss": 1.1879772003959208, | |
| "train_runtime": 144.3943, | |
| "train_samples_per_second": 18.837, | |
| "train_steps_per_second": 2.355 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 340, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 2.1078954658234368e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |