| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.04081632653061224, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0008163265306122449, | |
| "grad_norm": 0.7544206976890564, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5845, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008163265306122449, | |
| "eval_loss": 0.6241262555122375, | |
| "eval_runtime": 303.7147, | |
| "eval_samples_per_second": 3.398, | |
| "eval_steps_per_second": 1.699, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016326530612244899, | |
| "grad_norm": 0.5853733420372009, | |
| "learning_rate": 4e-05, | |
| "loss": 0.5365, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0024489795918367346, | |
| "grad_norm": 0.6549996733665466, | |
| "learning_rate": 6e-05, | |
| "loss": 0.6425, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0032653061224489797, | |
| "grad_norm": 0.614408552646637, | |
| "learning_rate": 8e-05, | |
| "loss": 0.5795, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.004081632653061225, | |
| "grad_norm": 0.5475334525108337, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5454, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.004081632653061225, | |
| "eval_loss": 0.5514864325523376, | |
| "eval_runtime": 306.4705, | |
| "eval_samples_per_second": 3.367, | |
| "eval_steps_per_second": 1.684, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.004897959183673469, | |
| "grad_norm": 0.6220462918281555, | |
| "learning_rate": 0.00012, | |
| "loss": 0.5158, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005714285714285714, | |
| "grad_norm": 0.6376144289970398, | |
| "learning_rate": 0.00014, | |
| "loss": 0.4644, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.006530612244897959, | |
| "grad_norm": 0.7968347668647766, | |
| "learning_rate": 0.00016, | |
| "loss": 0.4495, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.007346938775510204, | |
| "grad_norm": 0.6648410558700562, | |
| "learning_rate": 0.00018, | |
| "loss": 0.4173, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00816326530612245, | |
| "grad_norm": 0.8778969645500183, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4878, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00816326530612245, | |
| "eval_loss": 0.43802592158317566, | |
| "eval_runtime": 306.6738, | |
| "eval_samples_per_second": 3.365, | |
| "eval_steps_per_second": 1.683, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008979591836734694, | |
| "grad_norm": 0.7390214800834656, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 0.4016, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.009795918367346938, | |
| "grad_norm": 0.5307807922363281, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.439, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.010612244897959184, | |
| "grad_norm": 0.515744686126709, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.37, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.011428571428571429, | |
| "grad_norm": 0.5767286419868469, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.3815, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.012244897959183673, | |
| "grad_norm": 0.4507094919681549, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.3675, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.012244897959183673, | |
| "eval_loss": 0.3925292193889618, | |
| "eval_runtime": 306.4671, | |
| "eval_samples_per_second": 3.367, | |
| "eval_steps_per_second": 1.684, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.013061224489795919, | |
| "grad_norm": 0.45942962169647217, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.4134, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.013877551020408163, | |
| "grad_norm": 0.4870969355106354, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.389, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.014693877551020407, | |
| "grad_norm": 0.4066920876502991, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.4022, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.015510204081632653, | |
| "grad_norm": 0.44875219464302063, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 0.3507, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0163265306122449, | |
| "grad_norm": 0.49346503615379333, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.3737, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0163265306122449, | |
| "eval_loss": 0.38162362575531006, | |
| "eval_runtime": 306.1047, | |
| "eval_samples_per_second": 3.371, | |
| "eval_steps_per_second": 1.686, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.017142857142857144, | |
| "grad_norm": 0.42135682702064514, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.3401, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.017959183673469388, | |
| "grad_norm": 0.45447424054145813, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.3611, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.018775510204081632, | |
| "grad_norm": 0.4188750982284546, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.2922, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.019591836734693877, | |
| "grad_norm": 0.5015071630477905, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.3682, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.02040816326530612, | |
| "grad_norm": 0.4825016260147095, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.367, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.02040816326530612, | |
| "eval_loss": 0.3707207441329956, | |
| "eval_runtime": 306.4625, | |
| "eval_samples_per_second": 3.367, | |
| "eval_steps_per_second": 1.684, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.02122448979591837, | |
| "grad_norm": 0.45318421721458435, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.331, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.022040816326530613, | |
| "grad_norm": 0.4632524251937866, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.4417, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.022857142857142857, | |
| "grad_norm": 0.4687122404575348, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.3466, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0236734693877551, | |
| "grad_norm": 0.41113656759262085, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.3353, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.024489795918367346, | |
| "grad_norm": 0.41863101720809937, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3414, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.024489795918367346, | |
| "eval_loss": 0.36513009667396545, | |
| "eval_runtime": 306.2051, | |
| "eval_samples_per_second": 3.37, | |
| "eval_steps_per_second": 1.685, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.025306122448979593, | |
| "grad_norm": 0.4296859800815582, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.411, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.026122448979591838, | |
| "grad_norm": 0.3878820538520813, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.2489, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.026938775510204082, | |
| "grad_norm": 0.4258907437324524, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.3211, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.027755102040816326, | |
| "grad_norm": 0.5517330169677734, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.4105, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.02857142857142857, | |
| "grad_norm": 0.4672947824001312, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.3589, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.02857142857142857, | |
| "eval_loss": 0.3595900535583496, | |
| "eval_runtime": 306.5235, | |
| "eval_samples_per_second": 3.367, | |
| "eval_steps_per_second": 1.683, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.029387755102040815, | |
| "grad_norm": 0.3918313682079315, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.314, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.030204081632653063, | |
| "grad_norm": 0.40090271830558777, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.3451, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.031020408163265307, | |
| "grad_norm": 0.4959544241428375, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.3831, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.03183673469387755, | |
| "grad_norm": 0.44812849164009094, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.4219, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0326530612244898, | |
| "grad_norm": 0.5515211820602417, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.5255, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0326530612244898, | |
| "eval_loss": 0.3551081120967865, | |
| "eval_runtime": 305.9895, | |
| "eval_samples_per_second": 3.373, | |
| "eval_steps_per_second": 1.686, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03346938775510204, | |
| "grad_norm": 0.5151244401931763, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.3823, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.03428571428571429, | |
| "grad_norm": 0.4395885765552521, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.3138, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.03510204081632653, | |
| "grad_norm": 0.3345911204814911, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.3134, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.035918367346938776, | |
| "grad_norm": 0.44604966044425964, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.3128, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.036734693877551024, | |
| "grad_norm": 0.40180206298828125, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.3261, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.036734693877551024, | |
| "eval_loss": 0.355383962392807, | |
| "eval_runtime": 306.185, | |
| "eval_samples_per_second": 3.371, | |
| "eval_steps_per_second": 1.685, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.037551020408163265, | |
| "grad_norm": 0.35214531421661377, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.2845, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.03836734693877551, | |
| "grad_norm": 0.4088071286678314, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 0.2965, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.03918367346938775, | |
| "grad_norm": 0.38135460019111633, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.3156, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.4034852683544159, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 0.3078, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.04081632653061224, | |
| "grad_norm": 0.41931870579719543, | |
| "learning_rate": 0.0, | |
| "loss": 0.3163, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04081632653061224, | |
| "eval_loss": 0.35439395904541016, | |
| "eval_runtime": 306.6971, | |
| "eval_samples_per_second": 3.365, | |
| "eval_steps_per_second": 1.682, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.707288333503693e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |