| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06552006552006552, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0013104013104013104, | |
| "grad_norm": 48.18488693237305, | |
| "learning_rate": 2e-05, | |
| "loss": 21.5286, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0013104013104013104, | |
| "eval_loss": 2.674520969390869, | |
| "eval_runtime": 127.1923, | |
| "eval_samples_per_second": 5.055, | |
| "eval_steps_per_second": 2.532, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002620802620802621, | |
| "grad_norm": 43.056209564208984, | |
| "learning_rate": 4e-05, | |
| "loss": 21.6371, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.003931203931203931, | |
| "grad_norm": 47.93461608886719, | |
| "learning_rate": 6e-05, | |
| "loss": 19.8922, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.005241605241605242, | |
| "grad_norm": 47.43315124511719, | |
| "learning_rate": 8e-05, | |
| "loss": 22.5069, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.006552006552006552, | |
| "grad_norm": 42.69608688354492, | |
| "learning_rate": 0.0001, | |
| "loss": 17.1638, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.006552006552006552, | |
| "eval_loss": 1.6987773180007935, | |
| "eval_runtime": 128.1734, | |
| "eval_samples_per_second": 5.017, | |
| "eval_steps_per_second": 2.512, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.007862407862407862, | |
| "grad_norm": 37.23678970336914, | |
| "learning_rate": 0.00012, | |
| "loss": 13.1718, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.009172809172809172, | |
| "grad_norm": 32.32421112060547, | |
| "learning_rate": 0.00014, | |
| "loss": 9.7565, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.010483210483210484, | |
| "grad_norm": 19.69174575805664, | |
| "learning_rate": 0.00016, | |
| "loss": 5.35, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.011793611793611793, | |
| "grad_norm": 17.225276947021484, | |
| "learning_rate": 0.00018, | |
| "loss": 3.7587, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.013104013104013105, | |
| "grad_norm": 9.852200508117676, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8049, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.013104013104013105, | |
| "eval_loss": 0.22797086834907532, | |
| "eval_runtime": 128.169, | |
| "eval_samples_per_second": 5.017, | |
| "eval_steps_per_second": 2.512, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.014414414414414415, | |
| "grad_norm": 30.736356735229492, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 1.8166, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.015724815724815724, | |
| "grad_norm": 7.484240531921387, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.3285, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.017035217035217036, | |
| "grad_norm": 6.41568660736084, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.0993, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.018345618345618344, | |
| "grad_norm": 29.169870376586914, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.4899, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.019656019656019656, | |
| "grad_norm": 0.07240833342075348, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.0022, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.019656019656019656, | |
| "eval_loss": 0.00263240747153759, | |
| "eval_runtime": 128.3634, | |
| "eval_samples_per_second": 5.009, | |
| "eval_steps_per_second": 2.509, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.020966420966420967, | |
| "grad_norm": 0.03540218994021416, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.0009, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.022276822276822275, | |
| "grad_norm": 0.08068729937076569, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.0007, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.023587223587223587, | |
| "grad_norm": 0.11995162069797516, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.0012, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.024897624897624898, | |
| "grad_norm": 0.2867644131183624, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 0.001, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.02620802620802621, | |
| "grad_norm": 3.858811616897583, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.0096, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02620802620802621, | |
| "eval_loss": 0.002191605046391487, | |
| "eval_runtime": 128.1559, | |
| "eval_samples_per_second": 5.017, | |
| "eval_steps_per_second": 2.513, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.027518427518427518, | |
| "grad_norm": 0.03244931250810623, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.0002, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.02882882882882883, | |
| "grad_norm": 7.549134731292725, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.0246, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.03013923013923014, | |
| "grad_norm": 0.12704536318778992, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.0003, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.03144963144963145, | |
| "grad_norm": 0.42429715394973755, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.0009, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03276003276003276, | |
| "grad_norm": 0.11962774395942688, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.0005, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03276003276003276, | |
| "eval_loss": 0.00145976641215384, | |
| "eval_runtime": 128.2321, | |
| "eval_samples_per_second": 5.014, | |
| "eval_steps_per_second": 2.511, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03407043407043407, | |
| "grad_norm": 0.0005695931613445282, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.03538083538083538, | |
| "grad_norm": 0.0009148366516456008, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03669123669123669, | |
| "grad_norm": 0.00078869384014979, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.038001638001638, | |
| "grad_norm": 0.0009283472318202257, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.0, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.03931203931203931, | |
| "grad_norm": 0.0008322462672367692, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03931203931203931, | |
| "eval_loss": 0.0021525316406041384, | |
| "eval_runtime": 128.4235, | |
| "eval_samples_per_second": 5.007, | |
| "eval_steps_per_second": 2.507, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04062244062244062, | |
| "grad_norm": 0.00037751407944597304, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.0, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.041932841932841934, | |
| "grad_norm": 0.0007581758545711637, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.0, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.043243243243243246, | |
| "grad_norm": 0.0010002865456044674, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.04455364455364455, | |
| "grad_norm": 0.0007729699718765914, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.04586404586404586, | |
| "grad_norm": 0.001885723671875894, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.04586404586404586, | |
| "eval_loss": 0.0022208159789443016, | |
| "eval_runtime": 128.4374, | |
| "eval_samples_per_second": 5.006, | |
| "eval_steps_per_second": 2.507, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.04717444717444717, | |
| "grad_norm": 3.205965042114258, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.0101, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.048484848484848485, | |
| "grad_norm": 0.0007124874973669648, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.0, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.049795249795249796, | |
| "grad_norm": 23.423662185668945, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.039, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.05110565110565111, | |
| "grad_norm": 0.0009878461714833975, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.05241605241605242, | |
| "grad_norm": 0.00045861193211749196, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05241605241605242, | |
| "eval_loss": 0.0011409703874960542, | |
| "eval_runtime": 128.2757, | |
| "eval_samples_per_second": 5.013, | |
| "eval_steps_per_second": 2.51, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.053726453726453724, | |
| "grad_norm": 0.0006661771330982447, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.0, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.055036855036855035, | |
| "grad_norm": 0.001363381976261735, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.05634725634725635, | |
| "grad_norm": 0.0007221851265057921, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.0, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.05765765765765766, | |
| "grad_norm": 0.0014367022085934877, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.0, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.05896805896805897, | |
| "grad_norm": 0.0027935144025832415, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05896805896805897, | |
| "eval_loss": 0.0007023880025371909, | |
| "eval_runtime": 128.4557, | |
| "eval_samples_per_second": 5.006, | |
| "eval_steps_per_second": 2.507, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06027846027846028, | |
| "grad_norm": 0.0010182298719882965, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.061588861588861586, | |
| "grad_norm": 0.0728151947259903, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 0.0001, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0628992628992629, | |
| "grad_norm": 0.0016225130530074239, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.06420966420966422, | |
| "grad_norm": 0.0006315704667940736, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 0.0, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.06552006552006552, | |
| "grad_norm": 0.0017758873291313648, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06552006552006552, | |
| "eval_loss": 0.0005581722944043577, | |
| "eval_runtime": 128.3059, | |
| "eval_samples_per_second": 5.011, | |
| "eval_steps_per_second": 2.51, | |
| "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": 6.53458201903104e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |