| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.8808777429467085, | |
| "eval_steps": 50, | |
| "global_step": 1200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07836990595611286, | |
| "grad_norm": 0.9672619104385376, | |
| "learning_rate": 2.413127413127413e-05, | |
| "loss": 0.5333, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07836990595611286, | |
| "eval_loss": 0.21812455356121063, | |
| "eval_runtime": 136.5294, | |
| "eval_samples_per_second": 5.23, | |
| "eval_steps_per_second": 0.659, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15673981191222572, | |
| "grad_norm": 0.8239838480949402, | |
| "learning_rate": 2.3166023166023168e-05, | |
| "loss": 0.1948, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15673981191222572, | |
| "eval_loss": 0.180569127202034, | |
| "eval_runtime": 136.8617, | |
| "eval_samples_per_second": 5.217, | |
| "eval_steps_per_second": 0.658, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.23510971786833856, | |
| "grad_norm": 0.4985473155975342, | |
| "learning_rate": 2.2200772200772202e-05, | |
| "loss": 0.1772, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.23510971786833856, | |
| "eval_loss": 0.17201179265975952, | |
| "eval_runtime": 136.9237, | |
| "eval_samples_per_second": 5.215, | |
| "eval_steps_per_second": 0.657, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.31347962382445144, | |
| "grad_norm": 0.5671184659004211, | |
| "learning_rate": 2.1235521235521236e-05, | |
| "loss": 0.1674, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.31347962382445144, | |
| "eval_loss": 0.16670115292072296, | |
| "eval_runtime": 137.1006, | |
| "eval_samples_per_second": 5.208, | |
| "eval_steps_per_second": 0.656, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.39184952978056425, | |
| "grad_norm": 0.5542048215866089, | |
| "learning_rate": 2.0270270270270273e-05, | |
| "loss": 0.1639, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.39184952978056425, | |
| "eval_loss": 0.16387563943862915, | |
| "eval_runtime": 136.9694, | |
| "eval_samples_per_second": 5.213, | |
| "eval_steps_per_second": 0.657, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.4702194357366771, | |
| "grad_norm": 0.5355677604675293, | |
| "learning_rate": 1.9305019305019306e-05, | |
| "loss": 0.1603, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4702194357366771, | |
| "eval_loss": 0.16061373054981232, | |
| "eval_runtime": 136.8175, | |
| "eval_samples_per_second": 5.219, | |
| "eval_steps_per_second": 0.658, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.54858934169279, | |
| "grad_norm": 0.648248553276062, | |
| "learning_rate": 1.833976833976834e-05, | |
| "loss": 0.1563, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.54858934169279, | |
| "eval_loss": 0.1591891348361969, | |
| "eval_runtime": 136.7712, | |
| "eval_samples_per_second": 5.22, | |
| "eval_steps_per_second": 0.658, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6269592476489029, | |
| "grad_norm": 0.4232325851917267, | |
| "learning_rate": 1.7374517374517377e-05, | |
| "loss": 0.1595, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6269592476489029, | |
| "eval_loss": 0.15778718888759613, | |
| "eval_runtime": 137.2186, | |
| "eval_samples_per_second": 5.203, | |
| "eval_steps_per_second": 0.656, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7053291536050157, | |
| "grad_norm": 0.4626815915107727, | |
| "learning_rate": 1.640926640926641e-05, | |
| "loss": 0.1508, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.7053291536050157, | |
| "eval_loss": 0.1561730057001114, | |
| "eval_runtime": 136.858, | |
| "eval_samples_per_second": 5.217, | |
| "eval_steps_per_second": 0.658, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.7836990595611285, | |
| "grad_norm": 0.4221360385417938, | |
| "learning_rate": 1.5444015444015444e-05, | |
| "loss": 0.1581, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7836990595611285, | |
| "eval_loss": 0.15424229204654694, | |
| "eval_runtime": 137.1103, | |
| "eval_samples_per_second": 5.207, | |
| "eval_steps_per_second": 0.656, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8620689655172413, | |
| "grad_norm": 0.3882145583629608, | |
| "learning_rate": 1.4478764478764478e-05, | |
| "loss": 0.1441, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.8620689655172413, | |
| "eval_loss": 0.15357084572315216, | |
| "eval_runtime": 136.9739, | |
| "eval_samples_per_second": 5.213, | |
| "eval_steps_per_second": 0.657, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.9404388714733543, | |
| "grad_norm": 0.4727325141429901, | |
| "learning_rate": 1.3513513513513515e-05, | |
| "loss": 0.1487, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9404388714733543, | |
| "eval_loss": 0.15231835842132568, | |
| "eval_runtime": 136.9176, | |
| "eval_samples_per_second": 5.215, | |
| "eval_steps_per_second": 0.657, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0188087774294672, | |
| "grad_norm": 0.4438364803791046, | |
| "learning_rate": 1.2548262548262549e-05, | |
| "loss": 0.14, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.0188087774294672, | |
| "eval_loss": 0.15219952166080475, | |
| "eval_runtime": 137.0305, | |
| "eval_samples_per_second": 5.211, | |
| "eval_steps_per_second": 0.657, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.09717868338558, | |
| "grad_norm": 0.5468264818191528, | |
| "learning_rate": 1.1583011583011584e-05, | |
| "loss": 0.1366, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.09717868338558, | |
| "eval_loss": 0.1532224416732788, | |
| "eval_runtime": 137.4744, | |
| "eval_samples_per_second": 5.194, | |
| "eval_steps_per_second": 0.655, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1755485893416928, | |
| "grad_norm": 0.609539270401001, | |
| "learning_rate": 1.0617760617760618e-05, | |
| "loss": 0.1352, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.1755485893416928, | |
| "eval_loss": 0.15247634053230286, | |
| "eval_runtime": 137.6071, | |
| "eval_samples_per_second": 5.189, | |
| "eval_steps_per_second": 0.654, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.2539184952978055, | |
| "grad_norm": 0.5422230958938599, | |
| "learning_rate": 9.652509652509653e-06, | |
| "loss": 0.1278, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2539184952978055, | |
| "eval_loss": 0.15182016789913177, | |
| "eval_runtime": 137.443, | |
| "eval_samples_per_second": 5.195, | |
| "eval_steps_per_second": 0.655, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.3322884012539185, | |
| "grad_norm": 0.574143648147583, | |
| "learning_rate": 8.687258687258689e-06, | |
| "loss": 0.1315, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.3322884012539185, | |
| "eval_loss": 0.15190598368644714, | |
| "eval_runtime": 137.5049, | |
| "eval_samples_per_second": 5.193, | |
| "eval_steps_per_second": 0.655, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.4106583072100314, | |
| "grad_norm": 0.5425278544425964, | |
| "learning_rate": 7.722007722007722e-06, | |
| "loss": 0.1274, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.4106583072100314, | |
| "eval_loss": 0.15125001966953278, | |
| "eval_runtime": 137.6226, | |
| "eval_samples_per_second": 5.188, | |
| "eval_steps_per_second": 0.654, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.489028213166144, | |
| "grad_norm": 0.48568734526634216, | |
| "learning_rate": 6.7567567567567575e-06, | |
| "loss": 0.1347, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.489028213166144, | |
| "eval_loss": 0.15007726848125458, | |
| "eval_runtime": 137.6801, | |
| "eval_samples_per_second": 5.186, | |
| "eval_steps_per_second": 0.654, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.567398119122257, | |
| "grad_norm": 0.5619414448738098, | |
| "learning_rate": 5.791505791505792e-06, | |
| "loss": 0.1332, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.567398119122257, | |
| "eval_loss": 0.15013186633586884, | |
| "eval_runtime": 137.5165, | |
| "eval_samples_per_second": 5.192, | |
| "eval_steps_per_second": 0.654, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.64576802507837, | |
| "grad_norm": 0.5540875196456909, | |
| "learning_rate": 4.8262548262548266e-06, | |
| "loss": 0.1263, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.64576802507837, | |
| "eval_loss": 0.14901334047317505, | |
| "eval_runtime": 137.5107, | |
| "eval_samples_per_second": 5.192, | |
| "eval_steps_per_second": 0.654, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.7241379310344827, | |
| "grad_norm": 0.5203628540039062, | |
| "learning_rate": 3.861003861003861e-06, | |
| "loss": 0.1291, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.7241379310344827, | |
| "eval_loss": 0.14880172908306122, | |
| "eval_runtime": 137.6519, | |
| "eval_samples_per_second": 5.187, | |
| "eval_steps_per_second": 0.654, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.8025078369905956, | |
| "grad_norm": 0.5150081515312195, | |
| "learning_rate": 2.895752895752896e-06, | |
| "loss": 0.1287, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.8025078369905956, | |
| "eval_loss": 0.14855234324932098, | |
| "eval_runtime": 137.7759, | |
| "eval_samples_per_second": 5.182, | |
| "eval_steps_per_second": 0.653, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.8808777429467085, | |
| "grad_norm": 0.5428478717803955, | |
| "learning_rate": 1.9305019305019305e-06, | |
| "loss": 0.1278, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.8808777429467085, | |
| "eval_loss": 0.14838644862174988, | |
| "eval_runtime": 137.7999, | |
| "eval_samples_per_second": 5.181, | |
| "eval_steps_per_second": 0.653, | |
| "step": 1200 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 1300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.1478967485661184e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |