| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9900497512437811, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01990049751243781, | |
| "eval_loss": 1.5809879302978516, | |
| "eval_runtime": 5.5137, | |
| "eval_samples_per_second": 15.416, | |
| "eval_steps_per_second": 1.995, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05970149253731343, | |
| "grad_norm": 0.7845205068588257, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.5216, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.11940298507462686, | |
| "grad_norm": 0.8492766618728638, | |
| "learning_rate": 3e-05, | |
| "loss": 1.594, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.1791044776119403, | |
| "grad_norm": 0.899258553981781, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.5144, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1791044776119403, | |
| "eval_loss": 1.5270332098007202, | |
| "eval_runtime": 5.5991, | |
| "eval_samples_per_second": 15.181, | |
| "eval_steps_per_second": 1.965, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.23880597014925373, | |
| "grad_norm": 1.0086854696273804, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 1.43, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.29850746268656714, | |
| "grad_norm": 0.8446040749549866, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 1.3627, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.3582089552238806, | |
| "grad_norm": 0.7049321532249451, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 1.2934, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3582089552238806, | |
| "eval_loss": 1.2862956523895264, | |
| "eval_runtime": 5.6451, | |
| "eval_samples_per_second": 15.057, | |
| "eval_steps_per_second": 1.949, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.417910447761194, | |
| "grad_norm": 0.6026838421821594, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 1.2678, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.47761194029850745, | |
| "grad_norm": 0.6281826496124268, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 1.2053, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5373134328358209, | |
| "grad_norm": 0.626505434513092, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 1.1997, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5373134328358209, | |
| "eval_loss": 1.1615190505981445, | |
| "eval_runtime": 5.6361, | |
| "eval_samples_per_second": 15.081, | |
| "eval_steps_per_second": 1.952, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5970149253731343, | |
| "grad_norm": 0.5600003600120544, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 1.1329, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6567164179104478, | |
| "grad_norm": 0.554161787033081, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 1.0888, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.7164179104477612, | |
| "grad_norm": 0.5243486166000366, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 1.0326, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.7164179104477612, | |
| "eval_loss": 1.0722781419754028, | |
| "eval_runtime": 5.6371, | |
| "eval_samples_per_second": 15.079, | |
| "eval_steps_per_second": 1.951, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.7761194029850746, | |
| "grad_norm": 0.5589205026626587, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 1.0332, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.835820895522388, | |
| "grad_norm": 0.5794256329536438, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 1.0239, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.8955223880597015, | |
| "grad_norm": 0.5618190169334412, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.9851, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.8955223880597015, | |
| "eval_loss": 1.0372254848480225, | |
| "eval_runtime": 5.6453, | |
| "eval_samples_per_second": 15.057, | |
| "eval_steps_per_second": 1.949, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.9552238805970149, | |
| "grad_norm": 0.5606002807617188, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 1.0615, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.0149253731343284, | |
| "grad_norm": 1.186905860900879, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 1.1214, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.0746268656716418, | |
| "grad_norm": 0.5500369668006897, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.8825, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.0746268656716418, | |
| "eval_loss": 1.0109927654266357, | |
| "eval_runtime": 5.6371, | |
| "eval_samples_per_second": 15.079, | |
| "eval_steps_per_second": 1.951, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.1343283582089552, | |
| "grad_norm": 0.6391798853874207, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.9475, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.1940298507462686, | |
| "grad_norm": 0.5408726334571838, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.9773, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.2537313432835822, | |
| "grad_norm": 0.5797409415245056, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.9214, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.2537313432835822, | |
| "eval_loss": 0.9961395263671875, | |
| "eval_runtime": 5.6389, | |
| "eval_samples_per_second": 15.074, | |
| "eval_steps_per_second": 1.951, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.3134328358208955, | |
| "grad_norm": 0.5694305300712585, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.9053, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.373134328358209, | |
| "grad_norm": 0.5325923562049866, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.922, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.4328358208955223, | |
| "grad_norm": 0.6161549091339111, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.9302, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.4328358208955223, | |
| "eval_loss": 0.9865135550498962, | |
| "eval_runtime": 5.6362, | |
| "eval_samples_per_second": 15.081, | |
| "eval_steps_per_second": 1.952, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.4925373134328357, | |
| "grad_norm": 0.6062175035476685, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.9556, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.5522388059701493, | |
| "grad_norm": 0.6155171990394592, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 0.9589, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.6119402985074627, | |
| "grad_norm": 0.5605164766311646, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 0.8919, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.6119402985074627, | |
| "eval_loss": 0.9810183644294739, | |
| "eval_runtime": 5.6412, | |
| "eval_samples_per_second": 15.068, | |
| "eval_steps_per_second": 1.95, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.671641791044776, | |
| "grad_norm": 0.6275697350502014, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.9247, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.7313432835820897, | |
| "grad_norm": 0.5964632630348206, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.9138, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.7910447761194028, | |
| "grad_norm": 0.5972414016723633, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.8821, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.7910447761194028, | |
| "eval_loss": 0.9789036512374878, | |
| "eval_runtime": 5.6422, | |
| "eval_samples_per_second": 15.065, | |
| "eval_steps_per_second": 1.95, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.8507462686567164, | |
| "grad_norm": 0.6307412385940552, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.9657, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.9104477611940298, | |
| "grad_norm": 0.5744690895080566, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.8921, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.9701492537313432, | |
| "grad_norm": 0.6007259488105774, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 0.9583, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.9701492537313432, | |
| "eval_loss": 0.9787780046463013, | |
| "eval_runtime": 5.6467, | |
| "eval_samples_per_second": 15.053, | |
| "eval_steps_per_second": 1.948, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3065079924300186e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |