| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.35587188612099646, | |
| "eval_steps": 4, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0071174377224199285, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0071174377224199285, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.9332, | |
| "eval_samples_per_second": 29.875, | |
| "eval_steps_per_second": 3.782, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.014234875444839857, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.021352313167259787, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.028469750889679714, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.028469750889679714, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.2067, | |
| "eval_samples_per_second": 32.886, | |
| "eval_steps_per_second": 4.163, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.03558718861209965, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.042704626334519574, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0498220640569395, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.05693950177935943, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.05693950177935943, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.1934, | |
| "eval_samples_per_second": 32.947, | |
| "eval_steps_per_second": 4.171, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.06405693950177936, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0711743772241993, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07829181494661921, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.08540925266903915, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.08540925266903915, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.1961, | |
| "eval_samples_per_second": 32.935, | |
| "eval_steps_per_second": 4.169, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.09252669039145907, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.099644128113879, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.10676156583629894, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.11387900355871886, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.11387900355871886, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.1925, | |
| "eval_samples_per_second": 32.951, | |
| "eval_steps_per_second": 4.171, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.12099644128113879, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.12811387900355872, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.13523131672597866, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.1423487544483986, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1423487544483986, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.1928, | |
| "eval_samples_per_second": 32.95, | |
| "eval_steps_per_second": 4.171, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1494661921708185, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.15658362989323843, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.16370106761565836, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.0, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.1708185053380783, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.1708185053380783, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.1956, | |
| "eval_samples_per_second": 32.937, | |
| "eval_steps_per_second": 4.169, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.17793594306049823, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.18505338078291814, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.19217081850533807, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.199288256227758, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.199288256227758, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.1945, | |
| "eval_samples_per_second": 32.942, | |
| "eval_steps_per_second": 4.17, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.20640569395017794, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.0, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.21352313167259787, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2206405693950178, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.0, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.2277580071174377, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.0, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.2277580071174377, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.1955, | |
| "eval_samples_per_second": 32.937, | |
| "eval_steps_per_second": 4.169, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.23487544483985764, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.24199288256227758, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.2491103202846975, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.25622775800711745, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.25622775800711745, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.1951, | |
| "eval_samples_per_second": 32.939, | |
| "eval_steps_per_second": 4.169, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.26334519572953735, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.0, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.2704626334519573, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.0, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.2775800711743772, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.2846975088967972, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2846975088967972, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.1924, | |
| "eval_samples_per_second": 32.952, | |
| "eval_steps_per_second": 4.171, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2918149466192171, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.0, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.298932384341637, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.30604982206405695, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.0, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.31316725978647686, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.0, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.31316725978647686, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.1952, | |
| "eval_samples_per_second": 32.939, | |
| "eval_steps_per_second": 4.169, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.3202846975088968, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.3274021352313167, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.33451957295373663, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 0.0, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.3416370106761566, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.3416370106761566, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.1954, | |
| "eval_samples_per_second": 32.938, | |
| "eval_steps_per_second": 4.169, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.3487544483985765, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 0.0, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.35587188612099646, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 4, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.483774567120896e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |