| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.09551098376313276, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0019102196752626551, | |
| "grad_norm": 1.2359790802001953, | |
| "learning_rate": 5e-06, | |
| "loss": 2.9644, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0019102196752626551, | |
| "eval_loss": 3.0215418338775635, | |
| "eval_runtime": 14.0252, | |
| "eval_samples_per_second": 15.757, | |
| "eval_steps_per_second": 7.914, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0038204393505253103, | |
| "grad_norm": 1.0921162366867065, | |
| "learning_rate": 1e-05, | |
| "loss": 3.1595, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0057306590257879654, | |
| "grad_norm": 1.1526259183883667, | |
| "learning_rate": 1.5e-05, | |
| "loss": 3.0618, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.007640878701050621, | |
| "grad_norm": 1.199513554573059, | |
| "learning_rate": 2e-05, | |
| "loss": 3.1103, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.009551098376313277, | |
| "grad_norm": 1.2626410722732544, | |
| "learning_rate": 2.5e-05, | |
| "loss": 2.8805, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.011461318051575931, | |
| "grad_norm": 1.0098702907562256, | |
| "learning_rate": 3e-05, | |
| "loss": 3.1654, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.013371537726838587, | |
| "grad_norm": 1.0111432075500488, | |
| "learning_rate": 3.5e-05, | |
| "loss": 2.8632, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.015281757402101241, | |
| "grad_norm": 1.273777961730957, | |
| "learning_rate": 4e-05, | |
| "loss": 3.127, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.017191977077363897, | |
| "grad_norm": 1.0832494497299194, | |
| "learning_rate": 4.5e-05, | |
| "loss": 2.8716, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.019102196752626553, | |
| "grad_norm": 1.1544771194458008, | |
| "learning_rate": 5e-05, | |
| "loss": 3.0046, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.021012416427889206, | |
| "grad_norm": 1.253918170928955, | |
| "learning_rate": 4.99229333433282e-05, | |
| "loss": 2.9145, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.022922636103151862, | |
| "grad_norm": 1.0111428499221802, | |
| "learning_rate": 4.9692208514878444e-05, | |
| "loss": 2.9388, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.024832855778414518, | |
| "grad_norm": 1.001690149307251, | |
| "learning_rate": 4.9309248009941914e-05, | |
| "loss": 2.8392, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.024832855778414518, | |
| "eval_loss": 2.8127357959747314, | |
| "eval_runtime": 12.8181, | |
| "eval_samples_per_second": 17.241, | |
| "eval_steps_per_second": 8.66, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.026743075453677174, | |
| "grad_norm": 0.8471381068229675, | |
| "learning_rate": 4.877641290737884e-05, | |
| "loss": 2.6949, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.02865329512893983, | |
| "grad_norm": 0.766521692276001, | |
| "learning_rate": 4.8096988312782174e-05, | |
| "loss": 2.7781, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.030563514804202482, | |
| "grad_norm": 0.7984117865562439, | |
| "learning_rate": 4.72751631047092e-05, | |
| "loss": 2.76, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.03247373447946514, | |
| "grad_norm": 1.1073585748672485, | |
| "learning_rate": 4.6316004108852305e-05, | |
| "loss": 2.7066, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.034383954154727794, | |
| "grad_norm": 0.9880809187889099, | |
| "learning_rate": 4.522542485937369e-05, | |
| "loss": 2.7686, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03629417382999045, | |
| "grad_norm": 0.9087245464324951, | |
| "learning_rate": 4.401014914000078e-05, | |
| "loss": 2.7358, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.038204393505253106, | |
| "grad_norm": 1.1556546688079834, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 2.8673, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04011461318051576, | |
| "grad_norm": 0.7674556374549866, | |
| "learning_rate": 4.123620120825459e-05, | |
| "loss": 2.7329, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.04202483285577841, | |
| "grad_norm": 0.6524341106414795, | |
| "learning_rate": 3.969463130731183e-05, | |
| "loss": 2.5766, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.04393505253104107, | |
| "grad_norm": 0.6265513896942139, | |
| "learning_rate": 3.8062464117898724e-05, | |
| "loss": 2.6388, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.045845272206303724, | |
| "grad_norm": 0.572543740272522, | |
| "learning_rate": 3.634976249348867e-05, | |
| "loss": 2.5183, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.04775549188156638, | |
| "grad_norm": 0.5607729554176331, | |
| "learning_rate": 3.456708580912725e-05, | |
| "loss": 2.6954, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.049665711556829036, | |
| "grad_norm": 0.6020380258560181, | |
| "learning_rate": 3.272542485937369e-05, | |
| "loss": 2.6999, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.049665711556829036, | |
| "eval_loss": 2.671708106994629, | |
| "eval_runtime": 12.8491, | |
| "eval_samples_per_second": 17.2, | |
| "eval_steps_per_second": 8.639, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.05157593123209169, | |
| "grad_norm": 0.6852479577064514, | |
| "learning_rate": 3.083613409639764e-05, | |
| "loss": 2.748, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.05348615090735435, | |
| "grad_norm": 0.6889349222183228, | |
| "learning_rate": 2.8910861626005776e-05, | |
| "loss": 2.6797, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.055396370582617004, | |
| "grad_norm": 0.6189143061637878, | |
| "learning_rate": 2.6961477393196126e-05, | |
| "loss": 2.7193, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.05730659025787966, | |
| "grad_norm": 0.6611903309822083, | |
| "learning_rate": 2.5e-05, | |
| "loss": 2.6863, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.05921680993314231, | |
| "grad_norm": 0.5512831807136536, | |
| "learning_rate": 2.303852260680388e-05, | |
| "loss": 2.6735, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.061127029608404965, | |
| "grad_norm": 0.6554874777793884, | |
| "learning_rate": 2.1089138373994223e-05, | |
| "loss": 2.6688, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.06303724928366762, | |
| "grad_norm": 0.6053488850593567, | |
| "learning_rate": 1.9163865903602374e-05, | |
| "loss": 2.581, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.06494746895893028, | |
| "grad_norm": 0.5481260418891907, | |
| "learning_rate": 1.7274575140626318e-05, | |
| "loss": 2.6871, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.06685768863419293, | |
| "grad_norm": 0.6020957231521606, | |
| "learning_rate": 1.5432914190872757e-05, | |
| "loss": 2.5267, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.06876790830945559, | |
| "grad_norm": 0.5798554420471191, | |
| "learning_rate": 1.3650237506511331e-05, | |
| "loss": 2.5242, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.07067812798471824, | |
| "grad_norm": 0.6015373468399048, | |
| "learning_rate": 1.1937535882101281e-05, | |
| "loss": 2.6426, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0725883476599809, | |
| "grad_norm": 0.533322274684906, | |
| "learning_rate": 1.0305368692688174e-05, | |
| "loss": 2.7247, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.07449856733524356, | |
| "grad_norm": 0.5971359014511108, | |
| "learning_rate": 8.763798791745411e-06, | |
| "loss": 2.6084, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.07449856733524356, | |
| "eval_loss": 2.6242835521698, | |
| "eval_runtime": 12.8274, | |
| "eval_samples_per_second": 17.229, | |
| "eval_steps_per_second": 8.653, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.07640878701050621, | |
| "grad_norm": 0.5893213152885437, | |
| "learning_rate": 7.3223304703363135e-06, | |
| "loss": 2.6745, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07831900668576887, | |
| "grad_norm": 0.5584455728530884, | |
| "learning_rate": 5.989850859999227e-06, | |
| "loss": 2.6342, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.08022922636103152, | |
| "grad_norm": 0.5998557806015015, | |
| "learning_rate": 4.7745751406263165e-06, | |
| "loss": 2.4837, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.08213944603629417, | |
| "grad_norm": 0.5945307612419128, | |
| "learning_rate": 3.6839958911476957e-06, | |
| "loss": 2.5636, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.08404966571155682, | |
| "grad_norm": 0.6027778387069702, | |
| "learning_rate": 2.7248368952908053e-06, | |
| "loss": 2.8042, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.08595988538681948, | |
| "grad_norm": 0.549742579460144, | |
| "learning_rate": 1.9030116872178316e-06, | |
| "loss": 2.643, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.08787010506208214, | |
| "grad_norm": 0.5726300477981567, | |
| "learning_rate": 1.2235870926211619e-06, | |
| "loss": 2.6185, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.08978032473734479, | |
| "grad_norm": 0.5244706869125366, | |
| "learning_rate": 6.907519900580861e-07, | |
| "loss": 2.6166, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.09169054441260745, | |
| "grad_norm": 0.5466623902320862, | |
| "learning_rate": 3.077914851215585e-07, | |
| "loss": 2.558, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0936007640878701, | |
| "grad_norm": 0.5362265706062317, | |
| "learning_rate": 7.706665667180091e-08, | |
| "loss": 2.6386, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.09551098376313276, | |
| "grad_norm": 0.5353608727455139, | |
| "learning_rate": 0.0, | |
| "loss": 2.6433, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.84956424814592e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |