| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1261.8181818181818, | |
| "eval_steps": 500, | |
| "global_step": 3470, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 36.36363636363637, | |
| "grad_norm": 3.9305503368377686, | |
| "learning_rate": 0.00019992595626374085, | |
| "loss": 3.5508, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 72.72727272727273, | |
| "grad_norm": 7.143374919891357, | |
| "learning_rate": 0.00019969786478821292, | |
| "loss": 1.0142, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 109.0909090909091, | |
| "grad_norm": 4.481863498687744, | |
| "learning_rate": 0.00019932062382607466, | |
| "loss": 0.5034, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 145.45454545454547, | |
| "grad_norm": 5.447987079620361, | |
| "learning_rate": 0.00019878719501520854, | |
| "loss": 0.3536, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 181.8181818181818, | |
| "grad_norm": 6.030304431915283, | |
| "learning_rate": 0.00019810144350986773, | |
| "loss": 0.2281, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 218.1818181818182, | |
| "grad_norm": 3.293713331222534, | |
| "learning_rate": 0.0001972644266891692, | |
| "loss": 0.1648, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 254.54545454545453, | |
| "grad_norm": 3.844909191131592, | |
| "learning_rate": 0.0001962774351723822, | |
| "loss": 0.1083, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 290.90909090909093, | |
| "grad_norm": 6.67781400680542, | |
| "learning_rate": 0.00019514199082888708, | |
| "loss": 0.0708, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 327.27272727272725, | |
| "grad_norm": 0.4285012483596802, | |
| "learning_rate": 0.00019385984443156292, | |
| "loss": 0.0539, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 363.6363636363636, | |
| "grad_norm": 9.999969482421875, | |
| "learning_rate": 0.00019243297295722252, | |
| "loss": 0.0471, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 400.0, | |
| "grad_norm": 0.11577145010232925, | |
| "learning_rate": 0.00019086357653825758, | |
| "loss": 0.0361, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 436.3636363636364, | |
| "grad_norm": 0.04236361011862755, | |
| "learning_rate": 0.00018915407507019406, | |
| "loss": 0.0202, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 472.72727272727275, | |
| "grad_norm": 0.09598010778427124, | |
| "learning_rate": 0.0001873071044803886, | |
| "loss": 0.0204, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 509.09090909090907, | |
| "grad_norm": 0.04872431233525276, | |
| "learning_rate": 0.00018532551266361953, | |
| "loss": 0.0127, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 545.4545454545455, | |
| "grad_norm": 1.3734139204025269, | |
| "learning_rate": 0.00018321235509083966, | |
| "loss": 0.0125, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 581.8181818181819, | |
| "grad_norm": 0.058050643652677536, | |
| "learning_rate": 0.00018097089009786154, | |
| "loss": 0.0109, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 618.1818181818181, | |
| "grad_norm": 0.2065199464559555, | |
| "learning_rate": 0.0001786045738612397, | |
| "loss": 0.0054, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 654.5454545454545, | |
| "grad_norm": 0.011252381838858128, | |
| "learning_rate": 0.000176117055069097, | |
| "loss": 0.0092, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 690.9090909090909, | |
| "grad_norm": 0.2677134573459625, | |
| "learning_rate": 0.00017351216929511202, | |
| "loss": 0.0069, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 727.2727272727273, | |
| "grad_norm": 0.19302548468112946, | |
| "learning_rate": 0.00017079393308434222, | |
| "loss": 0.0052, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 763.6363636363636, | |
| "grad_norm": 0.09640073031187057, | |
| "learning_rate": 0.000167966537760003, | |
| "loss": 0.0056, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 800.0, | |
| "grad_norm": 0.05480790510773659, | |
| "learning_rate": 0.00016503434296075077, | |
| "loss": 0.0041, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 836.3636363636364, | |
| "grad_norm": 0.10284140706062317, | |
| "learning_rate": 0.00016200186991843633, | |
| "loss": 0.0026, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 872.7272727272727, | |
| "grad_norm": 0.03779178485274315, | |
| "learning_rate": 0.0001588737944866928, | |
| "loss": 0.0077, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 909.0909090909091, | |
| "grad_norm": 0.040165312588214874, | |
| "learning_rate": 0.00015565493993110856, | |
| "loss": 0.0068, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 945.4545454545455, | |
| "grad_norm": 0.0260086078196764, | |
| "learning_rate": 0.00015235026949210102, | |
| "loss": 0.004, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 981.8181818181819, | |
| "grad_norm": 0.0010075848549604416, | |
| "learning_rate": 0.0001489648787319599, | |
| "loss": 0.0029, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1018.1818181818181, | |
| "grad_norm": 0.1857517808675766, | |
| "learning_rate": 0.000145538952885112, | |
| "loss": 0.0025, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1054.5454545454545, | |
| "grad_norm": 0.009264905005693436, | |
| "learning_rate": 0.00014200857284118066, | |
| "loss": 0.0032, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1090.909090909091, | |
| "grad_norm": 0.005081487353891134, | |
| "learning_rate": 0.00013841341862277026, | |
| "loss": 0.0023, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1127.2727272727273, | |
| "grad_norm": 0.32677099108695984, | |
| "learning_rate": 0.0001347590336971037, | |
| "loss": 0.0055, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1163.6363636363637, | |
| "grad_norm": 0.014215439558029175, | |
| "learning_rate": 0.00013105105286086123, | |
| "loss": 0.0005, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1200.0, | |
| "grad_norm": 0.015997236594557762, | |
| "learning_rate": 0.00012729519355173254, | |
| "loss": 0.0009, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1236.3636363636363, | |
| "grad_norm": 0.668640673160553, | |
| "learning_rate": 0.00012349724703254215, | |
| "loss": 0.0024, | |
| "step": 3400 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 8000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4000, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0175353452509184e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |