| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.953846153846154, | |
| "eval_steps": 500, | |
| "global_step": 24, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12307692307692308, | |
| "grad_norm": 5.881288120029752, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.8683, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.24615384615384617, | |
| "grad_norm": 5.910057554919077, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 0.8728, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.36923076923076925, | |
| "grad_norm": 6.603049318418893, | |
| "learning_rate": 8e-05, | |
| "loss": 0.8714, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.49230769230769234, | |
| "grad_norm": 7.937989893518222, | |
| "learning_rate": 7.955323304900514e-05, | |
| "loss": 0.9348, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 6.666837918084534, | |
| "learning_rate": 7.822291223144564e-05, | |
| "loss": 0.8863, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.7384615384615385, | |
| "grad_norm": 4.617223284568761, | |
| "learning_rate": 7.603875471609677e-05, | |
| "loss": 0.8184, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.8615384615384616, | |
| "grad_norm": 7.202538725908792, | |
| "learning_rate": 7.30495509726398e-05, | |
| "loss": 0.8224, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.9846153846153847, | |
| "grad_norm": 3.7134305604831326, | |
| "learning_rate": 6.932207487319305e-05, | |
| "loss": 0.73, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.1076923076923078, | |
| "grad_norm": 5.77678057879716, | |
| "learning_rate": 6.493959207434934e-05, | |
| "loss": 1.3445, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 2.457296208076591, | |
| "learning_rate": 6.000000000000001e-05, | |
| "loss": 0.6753, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.353846153846154, | |
| "grad_norm": 1.7277702907502455, | |
| "learning_rate": 5.461364097465581e-05, | |
| "loss": 0.679, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.476923076923077, | |
| "grad_norm": 2.542271048846819, | |
| "learning_rate": 4.890083735825258e-05, | |
| "loss": 0.6444, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 1.2364293910648143, | |
| "learning_rate": 4.298920374345698e-05, | |
| "loss": 0.6291, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.7230769230769232, | |
| "grad_norm": 1.8800883331482068, | |
| "learning_rate": 3.7010796256543034e-05, | |
| "loss": 0.6202, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 1.1155652447895745, | |
| "learning_rate": 3.109916264174743e-05, | |
| "loss": 0.6044, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.9692307692307693, | |
| "grad_norm": 1.22973705104382, | |
| "learning_rate": 2.53863590253442e-05, | |
| "loss": 0.5969, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.0923076923076924, | |
| "grad_norm": 1.609651737996923, | |
| "learning_rate": 2.0000000000000012e-05, | |
| "loss": 1.1083, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.2153846153846155, | |
| "grad_norm": 0.6996543828925902, | |
| "learning_rate": 1.5060407925650662e-05, | |
| "loss": 0.5698, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.3384615384615386, | |
| "grad_norm": 0.8774794567467044, | |
| "learning_rate": 1.0677925126806956e-05, | |
| "loss": 0.5662, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 0.6833399495193899, | |
| "learning_rate": 6.950449027360213e-06, | |
| "loss": 0.5676, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.5846153846153848, | |
| "grad_norm": 0.39753644109612996, | |
| "learning_rate": 3.961245283903239e-06, | |
| "loss": 0.5574, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.707692307692308, | |
| "grad_norm": 0.3474450177854046, | |
| "learning_rate": 1.7770877685543687e-06, | |
| "loss": 0.5551, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.830769230769231, | |
| "grad_norm": 0.3762199558938556, | |
| "learning_rate": 4.467669509948591e-07, | |
| "loss": 0.5532, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.953846153846154, | |
| "grad_norm": 0.39774923662781286, | |
| "learning_rate": 0.0, | |
| "loss": 0.5554, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.953846153846154, | |
| "step": 24, | |
| "total_flos": 6.175904425354199e+17, | |
| "train_loss": 0.7346295664707819, | |
| "train_runtime": 5607.8463, | |
| "train_samples_per_second": 2.222, | |
| "train_steps_per_second": 0.004 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 24, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 6.175904425354199e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |