| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.2070393374741201, | |
| "eval_steps": 9, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004140786749482402, | |
| "grad_norm": 0.9301577806472778, | |
| "learning_rate": 1e-05, | |
| "loss": 3.1126, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004140786749482402, | |
| "eval_loss": 1.5828256607055664, | |
| "eval_runtime": 86.4747, | |
| "eval_samples_per_second": 2.359, | |
| "eval_steps_per_second": 0.301, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.008281573498964804, | |
| "grad_norm": 0.8852660655975342, | |
| "learning_rate": 2e-05, | |
| "loss": 3.2312, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.012422360248447204, | |
| "grad_norm": 0.9227669835090637, | |
| "learning_rate": 3e-05, | |
| "loss": 2.8183, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.016563146997929608, | |
| "grad_norm": 1.096196174621582, | |
| "learning_rate": 4e-05, | |
| "loss": 3.433, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.020703933747412008, | |
| "grad_norm": 1.1805181503295898, | |
| "learning_rate": 5e-05, | |
| "loss": 2.9417, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.024844720496894408, | |
| "grad_norm": 1.0887078046798706, | |
| "learning_rate": 6e-05, | |
| "loss": 2.7484, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.028985507246376812, | |
| "grad_norm": 1.1074923276901245, | |
| "learning_rate": 7e-05, | |
| "loss": 2.8224, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.033126293995859216, | |
| "grad_norm": 1.1030354499816895, | |
| "learning_rate": 8e-05, | |
| "loss": 2.7812, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.037267080745341616, | |
| "grad_norm": 1.2491376399993896, | |
| "learning_rate": 9e-05, | |
| "loss": 2.6374, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.037267080745341616, | |
| "eval_loss": 1.3686977624893188, | |
| "eval_runtime": 86.6236, | |
| "eval_samples_per_second": 2.355, | |
| "eval_steps_per_second": 0.3, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.041407867494824016, | |
| "grad_norm": 1.2215477228164673, | |
| "learning_rate": 0.0001, | |
| "loss": 2.9145, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.045548654244306416, | |
| "grad_norm": 1.2886161804199219, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 2.2524, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.049689440993788817, | |
| "grad_norm": 1.200274109840393, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 2.3977, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.053830227743271224, | |
| "grad_norm": 1.059577465057373, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 2.1013, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.057971014492753624, | |
| "grad_norm": 1.544952154159546, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 2.3457, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.062111801242236024, | |
| "grad_norm": 1.3836092948913574, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 2.7073, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.06625258799171843, | |
| "grad_norm": 1.410312533378601, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 2.8543, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.07039337474120083, | |
| "grad_norm": 1.2303005456924438, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 2.3382, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.07453416149068323, | |
| "grad_norm": 1.4924076795578003, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 2.2618, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07453416149068323, | |
| "eval_loss": 1.0939579010009766, | |
| "eval_runtime": 86.594, | |
| "eval_samples_per_second": 2.356, | |
| "eval_steps_per_second": 0.3, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07867494824016563, | |
| "grad_norm": 1.4655351638793945, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.5699, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.08281573498964803, | |
| "grad_norm": 1.2195974588394165, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 1.8441, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08695652173913043, | |
| "grad_norm": 1.2934242486953735, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.2081, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.09109730848861283, | |
| "grad_norm": 1.0533246994018555, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 2.0406, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.09523809523809523, | |
| "grad_norm": 1.3303511142730713, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 2.3889, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.09937888198757763, | |
| "grad_norm": 1.2019553184509277, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.6975, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.10351966873706005, | |
| "grad_norm": 1.1562660932540894, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 2.0118, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.10766045548654245, | |
| "grad_norm": 1.2384679317474365, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 1.8618, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.11180124223602485, | |
| "grad_norm": 1.2058842182159424, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 1.8744, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.11180124223602485, | |
| "eval_loss": 1.000207543373108, | |
| "eval_runtime": 86.613, | |
| "eval_samples_per_second": 2.355, | |
| "eval_steps_per_second": 0.3, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.11594202898550725, | |
| "grad_norm": 1.3206522464752197, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 2.4724, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.12008281573498965, | |
| "grad_norm": 1.1769299507141113, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 2.2312, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.12422360248447205, | |
| "grad_norm": 1.1039841175079346, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 1.8363, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12836438923395446, | |
| "grad_norm": 1.1677961349487305, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 2.0295, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.13250517598343686, | |
| "grad_norm": 1.136795163154602, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 1.7927, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.13664596273291926, | |
| "grad_norm": 1.1699427366256714, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 2.1008, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.14078674948240166, | |
| "grad_norm": 1.0956039428710938, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 1.8778, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.14492753623188406, | |
| "grad_norm": 1.106882929801941, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 1.8638, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.14906832298136646, | |
| "grad_norm": 1.176102638244629, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 2.0063, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.14906832298136646, | |
| "eval_loss": 0.9627537727355957, | |
| "eval_runtime": 86.6445, | |
| "eval_samples_per_second": 2.354, | |
| "eval_steps_per_second": 0.3, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.15320910973084886, | |
| "grad_norm": 1.2707808017730713, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2.0563, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.15734989648033126, | |
| "grad_norm": 1.1810678243637085, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 1.7766, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.16149068322981366, | |
| "grad_norm": 1.2528388500213623, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 2.2023, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.16563146997929606, | |
| "grad_norm": 1.1106541156768799, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 1.9607, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.16977225672877846, | |
| "grad_norm": 1.3797121047973633, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 2.2966, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 1.0759304761886597, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 1.9743, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.17805383022774326, | |
| "grad_norm": 1.485608696937561, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 2.1251, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.18219461697722567, | |
| "grad_norm": 1.1569021940231323, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 1.9108, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.18633540372670807, | |
| "grad_norm": 1.4364919662475586, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.561, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.18633540372670807, | |
| "eval_loss": 0.9237259030342102, | |
| "eval_runtime": 86.6431, | |
| "eval_samples_per_second": 2.354, | |
| "eval_steps_per_second": 0.3, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 1.126060128211975, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.8225, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.19461697722567287, | |
| "grad_norm": 1.137122392654419, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 1.8008, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.19875776397515527, | |
| "grad_norm": 1.1307929754257202, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 1.8732, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.2028985507246377, | |
| "grad_norm": 1.3419166803359985, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 1.782, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.2070393374741201, | |
| "grad_norm": 1.3455865383148193, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 1.4769, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1489572231839744e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |