| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.8, | |
| "eval_steps": 3, | |
| "global_step": 36, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.17093713581562042, | |
| "learning_rate": 2e-05, | |
| "loss": 1.4615, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 1.4899382591247559, | |
| "eval_runtime": 17.3107, | |
| "eval_samples_per_second": 5.777, | |
| "eval_steps_per_second": 2.888, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.19339510798454285, | |
| "learning_rate": 4e-05, | |
| "loss": 1.4241, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.1669788658618927, | |
| "learning_rate": 6e-05, | |
| "loss": 1.3846, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 1.4858685731887817, | |
| "eval_runtime": 17.4215, | |
| "eval_samples_per_second": 5.74, | |
| "eval_steps_per_second": 2.87, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.14142441749572754, | |
| "learning_rate": 8e-05, | |
| "loss": 1.2219, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.15717843174934387, | |
| "learning_rate": 0.0001, | |
| "loss": 1.3617, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.16309261322021484, | |
| "learning_rate": 0.00012, | |
| "loss": 1.3667, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 1.439871072769165, | |
| "eval_runtime": 17.6339, | |
| "eval_samples_per_second": 5.671, | |
| "eval_steps_per_second": 2.835, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.15078029036521912, | |
| "learning_rate": 0.00014, | |
| "loss": 1.3008, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.13603582978248596, | |
| "learning_rate": 0.00016, | |
| "loss": 1.3333, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.1104956567287445, | |
| "learning_rate": 0.00018, | |
| "loss": 1.267, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_loss": 1.3381670713424683, | |
| "eval_runtime": 17.2986, | |
| "eval_samples_per_second": 5.781, | |
| "eval_steps_per_second": 2.89, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.09913735836744308, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2946, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.11903145164251328, | |
| "learning_rate": 0.000199658449300667, | |
| "loss": 1.2921, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.11169299483299255, | |
| "learning_rate": 0.00019863613034027224, | |
| "loss": 1.2276, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 1.2940881252288818, | |
| "eval_runtime": 17.4061, | |
| "eval_samples_per_second": 5.745, | |
| "eval_steps_per_second": 2.873, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 0.1135605201125145, | |
| "learning_rate": 0.00019694002659393305, | |
| "loss": 1.2666, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 0.1120605319738388, | |
| "learning_rate": 0.00019458172417006347, | |
| "loss": 1.2589, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.10806083679199219, | |
| "learning_rate": 0.00019157733266550575, | |
| "loss": 1.2515, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_loss": 1.2792645692825317, | |
| "eval_runtime": 17.2617, | |
| "eval_samples_per_second": 5.793, | |
| "eval_steps_per_second": 2.897, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.09928147494792938, | |
| "learning_rate": 0.0001879473751206489, | |
| "loss": 1.1636, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 0.09012632817029953, | |
| "learning_rate": 0.00018371664782625287, | |
| "loss": 1.1601, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.09199394285678864, | |
| "learning_rate": 0.00017891405093963938, | |
| "loss": 1.2275, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_loss": 1.2555986642837524, | |
| "eval_runtime": 17.3997, | |
| "eval_samples_per_second": 5.747, | |
| "eval_steps_per_second": 2.874, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 0.096685491502285, | |
| "learning_rate": 0.00017357239106731317, | |
| "loss": 1.228, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.08053378760814667, | |
| "learning_rate": 0.00016772815716257412, | |
| "loss": 1.192, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 0.07304864376783371, | |
| "learning_rate": 0.0001614212712689668, | |
| "loss": 1.1351, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "eval_loss": 1.2347127199172974, | |
| "eval_runtime": 17.3396, | |
| "eval_samples_per_second": 5.767, | |
| "eval_steps_per_second": 2.884, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 0.0821811631321907, | |
| "learning_rate": 0.00015469481581224272, | |
| "loss": 1.2031, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.07574562728404999, | |
| "learning_rate": 0.00014759473930370736, | |
| "loss": 1.2476, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.07235240191221237, | |
| "learning_rate": 0.00014016954246529696, | |
| "loss": 1.2701, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "eval_loss": 1.2253003120422363, | |
| "eval_runtime": 17.292, | |
| "eval_samples_per_second": 5.783, | |
| "eval_steps_per_second": 2.892, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 0.07071871310472488, | |
| "learning_rate": 0.00013246994692046836, | |
| "loss": 1.2051, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 0.07875131815671921, | |
| "learning_rate": 0.00012454854871407994, | |
| "loss": 1.192, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.06468148529529572, | |
| "learning_rate": 0.00011645945902807341, | |
| "loss": 1.1487, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "eval_loss": 1.2213425636291504, | |
| "eval_runtime": 17.7102, | |
| "eval_samples_per_second": 5.646, | |
| "eval_steps_per_second": 2.823, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 0.0741707906126976, | |
| "learning_rate": 0.00010825793454723325, | |
| "loss": 1.1673, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.06802140176296234, | |
| "learning_rate": 0.0001, | |
| "loss": 1.2054, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 0.06834083795547485, | |
| "learning_rate": 9.174206545276677e-05, | |
| "loss": 1.1518, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "eval_loss": 1.220943808555603, | |
| "eval_runtime": 17.4872, | |
| "eval_samples_per_second": 5.718, | |
| "eval_steps_per_second": 2.859, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.06714992970228195, | |
| "learning_rate": 8.35405409719266e-05, | |
| "loss": 1.183, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 0.06744072586297989, | |
| "learning_rate": 7.54514512859201e-05, | |
| "loss": 1.2098, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.06815183162689209, | |
| "learning_rate": 6.753005307953167e-05, | |
| "loss": 1.1942, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_loss": 1.217085599899292, | |
| "eval_runtime": 17.2384, | |
| "eval_samples_per_second": 5.801, | |
| "eval_steps_per_second": 2.901, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 0.07002735882997513, | |
| "learning_rate": 5.983045753470308e-05, | |
| "loss": 1.1896, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 0.07221011817455292, | |
| "learning_rate": 5.240526069629265e-05, | |
| "loss": 1.1795, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.07009800523519516, | |
| "learning_rate": 4.530518418775733e-05, | |
| "loss": 1.1122, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_loss": 1.2147068977355957, | |
| "eval_runtime": 17.3366, | |
| "eval_samples_per_second": 5.768, | |
| "eval_steps_per_second": 2.884, | |
| "step": 36 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 48, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 12, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7474703266480128.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |