| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.007537519348542971, | |
| "eval_steps": 4, | |
| "global_step": 28, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00026919711959082036, | |
| "grad_norm": 7.305245876312256, | |
| "learning_rate": 2e-05, | |
| "loss": 6.0527, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00026919711959082036, | |
| "eval_loss": 6.004375457763672, | |
| "eval_runtime": 186.7841, | |
| "eval_samples_per_second": 8.379, | |
| "eval_steps_per_second": 4.192, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005383942391816407, | |
| "grad_norm": 6.773212432861328, | |
| "learning_rate": 4e-05, | |
| "loss": 6.151, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0008075913587724611, | |
| "grad_norm": 7.554437160491943, | |
| "learning_rate": 6e-05, | |
| "loss": 6.398, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0010767884783632815, | |
| "grad_norm": 6.408179759979248, | |
| "learning_rate": 8e-05, | |
| "loss": 5.7049, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0010767884783632815, | |
| "eval_loss": 5.3220696449279785, | |
| "eval_runtime": 187.1601, | |
| "eval_samples_per_second": 8.362, | |
| "eval_steps_per_second": 4.184, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0013459855979541019, | |
| "grad_norm": 6.113898754119873, | |
| "learning_rate": 0.0001, | |
| "loss": 5.5139, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0016151827175449223, | |
| "grad_norm": 4.3670973777771, | |
| "learning_rate": 0.00012, | |
| "loss": 4.8011, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0018843798371357427, | |
| "grad_norm": 8.779766082763672, | |
| "learning_rate": 0.00014, | |
| "loss": 4.9951, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.002153576956726563, | |
| "grad_norm": 8.467183113098145, | |
| "learning_rate": 0.00016, | |
| "loss": 4.8208, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.002153576956726563, | |
| "eval_loss": 4.797426700592041, | |
| "eval_runtime": 187.0275, | |
| "eval_samples_per_second": 8.368, | |
| "eval_steps_per_second": 4.187, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0024227740763173833, | |
| "grad_norm": 6.028713226318359, | |
| "learning_rate": 0.00018, | |
| "loss": 4.9807, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0026919711959082038, | |
| "grad_norm": 3.5283517837524414, | |
| "learning_rate": 0.0002, | |
| "loss": 4.4461, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.002961168315499024, | |
| "grad_norm": 3.948577880859375, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 4.0834, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0032303654350898446, | |
| "grad_norm": 3.531851053237915, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 4.6573, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0032303654350898446, | |
| "eval_loss": 4.350089073181152, | |
| "eval_runtime": 186.8405, | |
| "eval_samples_per_second": 8.376, | |
| "eval_steps_per_second": 4.191, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.003499562554680665, | |
| "grad_norm": 3.562119483947754, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 4.4314, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0037687596742714854, | |
| "grad_norm": 3.721893548965454, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 4.4538, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.004037956793862305, | |
| "grad_norm": 2.8506362438201904, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 4.2002, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.004307153913453126, | |
| "grad_norm": 2.9634995460510254, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 4.3413, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.004307153913453126, | |
| "eval_loss": 4.165503978729248, | |
| "eval_runtime": 187.0045, | |
| "eval_samples_per_second": 8.369, | |
| "eval_steps_per_second": 4.187, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.004576351033043946, | |
| "grad_norm": 3.4099552631378174, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 4.133, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.004845548152634767, | |
| "grad_norm": 3.074971914291382, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 3.9521, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.005114745272225587, | |
| "grad_norm": 2.5733466148376465, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 4.1701, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0053839423918164075, | |
| "grad_norm": 3.7706105709075928, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 3.9503, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0053839423918164075, | |
| "eval_loss": 4.0563483238220215, | |
| "eval_runtime": 187.0013, | |
| "eval_samples_per_second": 8.369, | |
| "eval_steps_per_second": 4.187, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.005653139511407228, | |
| "grad_norm": 2.7933688163757324, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 3.8037, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.005922336630998048, | |
| "grad_norm": 2.6206228733062744, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 4.1224, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.006191533750588869, | |
| "grad_norm": 2.956394910812378, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 3.8059, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.006460730870179689, | |
| "grad_norm": 2.445277690887451, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 3.8613, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.006460730870179689, | |
| "eval_loss": 3.9950814247131348, | |
| "eval_runtime": 186.8191, | |
| "eval_samples_per_second": 8.377, | |
| "eval_steps_per_second": 4.191, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.00672992798977051, | |
| "grad_norm": 2.6795260906219482, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 3.646, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.00699912510936133, | |
| "grad_norm": 2.925718307495117, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 4.1437, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0072683222289521504, | |
| "grad_norm": 2.81306529045105, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 3.9201, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.007537519348542971, | |
| "grad_norm": 2.4982492923736572, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 3.8325, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.007537519348542971, | |
| "eval_loss": 3.943394660949707, | |
| "eval_runtime": 187.1247, | |
| "eval_samples_per_second": 8.363, | |
| "eval_steps_per_second": 4.184, | |
| "step": 28 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 4, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.147304212319437e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |