| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.36542072025943, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.41249775886535645, | |
| "learning_rate": 0.00011363636363636365, | |
| "loss": 1.6991, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.4261733889579773, | |
| "learning_rate": 0.00019999118973067944, | |
| "loss": 1.463, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.46912631392478943, | |
| "learning_rate": 0.00019976490347667713, | |
| "loss": 1.4387, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.4052003026008606, | |
| "learning_rate": 0.00019923349759697913, | |
| "loss": 1.4552, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.41492465138435364, | |
| "learning_rate": 0.00019839859733610985, | |
| "loss": 1.432, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.4481758177280426, | |
| "learning_rate": 0.0001972627561416867, | |
| "loss": 1.4365, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.5057295560836792, | |
| "learning_rate": 0.00019582944785499073, | |
| "loss": 1.4398, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.4403720796108246, | |
| "learning_rate": 0.00019410305608661744, | |
| "loss": 1.4236, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.4727613627910614, | |
| "learning_rate": 0.00019208886080970187, | |
| "loss": 1.4109, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.5325965285301208, | |
| "learning_rate": 0.0001897930222117203, | |
| "loss": 1.4556, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.5341053605079651, | |
| "learning_rate": 0.00018722256185425655, | |
| "loss": 1.426, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.6165050268173218, | |
| "learning_rate": 0.00018438534119835363, | |
| "loss": 1.3777, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.455497682094574, | |
| "learning_rate": 0.0001812900375611279, | |
| "loss": 1.4352, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.4459982216358185, | |
| "learning_rate": 0.00017794611757718015, | |
| "loss": 1.3917, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.488208144903183, | |
| "learning_rate": 0.00017436380824596915, | |
| "loss": 1.3801, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.6411360502243042, | |
| "learning_rate": 0.00017055406565369532, | |
| "loss": 1.4015, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.6391683220863342, | |
| "learning_rate": 0.0001665285414653555, | |
| "loss": 1.3578, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.4634678363800049, | |
| "learning_rate": 0.00016229954728944895, | |
| "loss": 1.3958, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.5685352087020874, | |
| "learning_rate": 0.00015788001702432132, | |
| "loss": 1.3824, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.48762017488479614, | |
| "learning_rate": 0.0001532834673013053, | |
| "loss": 1.3651, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.4447900354862213, | |
| "learning_rate": 0.00014852395614563935, | |
| "loss": 1.3928, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.5502452850341797, | |
| "learning_rate": 0.00014361603998159388, | |
| "loss": 1.3901, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.5311463475227356, | |
| "learning_rate": 0.00013857472911330131, | |
| "loss": 1.3506, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.5501254796981812, | |
| "learning_rate": 0.0001334154418174456, | |
| "loss": 1.3611, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.6120589375495911, | |
| "learning_rate": 0.00012815395718821415, | |
| "loss": 1.3879, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.6142286658287048, | |
| "learning_rate": 0.0001228063668787309, | |
| "loss": 1.3633, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.610168993473053, | |
| "learning_rate": 0.00011738902588656242, | |
| "loss": 1.3522, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.5812483429908752, | |
| "learning_rate": 0.00011191850253381601, | |
| "loss": 1.3938, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.6008601784706116, | |
| "learning_rate": 0.00010641152779480806, | |
| "loss": 1.3396, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 0.6404789686203003, | |
| "learning_rate": 0.00010088494412627968, | |
| "loss": 1.3007, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 0.6050863862037659, | |
| "learning_rate": 9.535565395665562e-05, | |
| "loss": 1.2374, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 0.5933852195739746, | |
| "learning_rate": 8.984056799188677e-05, | |
| "loss": 1.2135, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 0.7690561413764954, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 1.2479, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.735683023929596, | |
| "learning_rate": 7.892038270437153e-05, | |
| "loss": 1.2115, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "grad_norm": 0.7068488597869873, | |
| "learning_rate": 7.35486815279806e-05, | |
| "loss": 1.2184, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 0.821847677230835, | |
| "learning_rate": 6.825787870471872e-05, | |
| "loss": 1.2054, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 0.7907322645187378, | |
| "learning_rate": 6.30641555540761e-05, | |
| "loss": 1.2289, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 0.6236514449119568, | |
| "learning_rate": 5.798339648839073e-05, | |
| "loss": 1.2113, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 0.6800610423088074, | |
| "learning_rate": 5.303114043217771e-05, | |
| "loss": 1.201, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 0.6555848121643066, | |
| "learning_rate": 4.8222533298093295e-05, | |
| "loss": 1.2028, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 2928, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "total_flos": 4.863950904700109e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |