| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.5483870967741935, | |
| "eval_steps": 16, | |
| "global_step": 48, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03225806451612903, | |
| "grad_norm": 12.588220596313477, | |
| "learning_rate": 2.0000000000000002e-07, | |
| "loss": 1.3146, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03225806451612903, | |
| "eval_loss": 1.3345524072647095, | |
| "eval_runtime": 19.3383, | |
| "eval_samples_per_second": 40.076, | |
| "eval_steps_per_second": 5.016, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06451612903225806, | |
| "grad_norm": 12.520931243896484, | |
| "learning_rate": 4.0000000000000003e-07, | |
| "loss": 1.3028, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0967741935483871, | |
| "grad_norm": 11.893294334411621, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 1.288, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.12903225806451613, | |
| "grad_norm": 12.266352653503418, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 1.2856, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.16129032258064516, | |
| "grad_norm": 12.19696044921875, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.3012, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.1935483870967742, | |
| "grad_norm": 10.874934196472168, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "loss": 1.2617, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.22580645161290322, | |
| "grad_norm": 11.033644676208496, | |
| "learning_rate": 1.4000000000000001e-06, | |
| "loss": 1.3117, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.25806451612903225, | |
| "grad_norm": 7.418929576873779, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 1.2131, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.2903225806451613, | |
| "grad_norm": 5.99593448638916, | |
| "learning_rate": 1.8000000000000001e-06, | |
| "loss": 1.2242, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3225806451612903, | |
| "grad_norm": 4.825831890106201, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.2185, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3548387096774194, | |
| "grad_norm": 5.234226226806641, | |
| "learning_rate": 2.2e-06, | |
| "loss": 1.1888, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.3870967741935484, | |
| "grad_norm": 6.353809356689453, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 1.1947, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.41935483870967744, | |
| "grad_norm": 6.419878005981445, | |
| "learning_rate": 2.6e-06, | |
| "loss": 1.1785, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.45161290322580644, | |
| "grad_norm": 5.540423393249512, | |
| "learning_rate": 2.8000000000000003e-06, | |
| "loss": 1.1535, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.4838709677419355, | |
| "grad_norm": 4.607199668884277, | |
| "learning_rate": 3e-06, | |
| "loss": 1.1567, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5161290322580645, | |
| "grad_norm": 4.273080825805664, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "loss": 1.1667, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5161290322580645, | |
| "eval_loss": 1.1729758977890015, | |
| "eval_runtime": 18.7332, | |
| "eval_samples_per_second": 41.37, | |
| "eval_steps_per_second": 5.178, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5483870967741935, | |
| "grad_norm": 3.5295605659484863, | |
| "learning_rate": 3.4000000000000005e-06, | |
| "loss": 1.1368, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.5806451612903226, | |
| "grad_norm": 2.8693783283233643, | |
| "learning_rate": 3.6000000000000003e-06, | |
| "loss": 1.1796, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.6129032258064516, | |
| "grad_norm": 2.4456138610839844, | |
| "learning_rate": 3.8000000000000005e-06, | |
| "loss": 1.0856, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.6451612903225806, | |
| "grad_norm": 1.8665717840194702, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.0991, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6774193548387096, | |
| "grad_norm": 2.3066937923431396, | |
| "learning_rate": 4.2000000000000004e-06, | |
| "loss": 1.0935, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.7096774193548387, | |
| "grad_norm": 2.1404623985290527, | |
| "learning_rate": 4.4e-06, | |
| "loss": 1.0889, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.7419354838709677, | |
| "grad_norm": 2.023366689682007, | |
| "learning_rate": 4.600000000000001e-06, | |
| "loss": 1.103, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.7741935483870968, | |
| "grad_norm": 2.192676544189453, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 1.0577, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8064516129032258, | |
| "grad_norm": 2.195632219314575, | |
| "learning_rate": 5e-06, | |
| "loss": 1.0778, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8387096774193549, | |
| "grad_norm": 1.9270845651626587, | |
| "learning_rate": 5.2e-06, | |
| "loss": 1.0823, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.8709677419354839, | |
| "grad_norm": 2.089909553527832, | |
| "learning_rate": 5.400000000000001e-06, | |
| "loss": 1.07, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.9032258064516129, | |
| "grad_norm": 1.8139009475708008, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 1.0714, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.9354838709677419, | |
| "grad_norm": 1.8297277688980103, | |
| "learning_rate": 5.8e-06, | |
| "loss": 1.1, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.967741935483871, | |
| "grad_norm": 2.1214663982391357, | |
| "learning_rate": 6e-06, | |
| "loss": 1.0327, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.7602064609527588, | |
| "learning_rate": 6.200000000000001e-06, | |
| "loss": 1.0699, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.032258064516129, | |
| "grad_norm": 2.081988573074341, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 0.9551, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.032258064516129, | |
| "eval_loss": 1.0734707117080688, | |
| "eval_runtime": 18.7246, | |
| "eval_samples_per_second": 41.389, | |
| "eval_steps_per_second": 5.18, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.064516129032258, | |
| "grad_norm": 1.8479481935501099, | |
| "learning_rate": 6.600000000000001e-06, | |
| "loss": 0.9424, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.096774193548387, | |
| "grad_norm": 1.8964370489120483, | |
| "learning_rate": 6.800000000000001e-06, | |
| "loss": 0.915, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.129032258064516, | |
| "grad_norm": 1.7067584991455078, | |
| "learning_rate": 7e-06, | |
| "loss": 0.8889, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.1612903225806452, | |
| "grad_norm": 1.877442717552185, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 0.8858, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.1935483870967742, | |
| "grad_norm": 1.8160911798477173, | |
| "learning_rate": 7.4e-06, | |
| "loss": 0.8999, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.2258064516129032, | |
| "grad_norm": 1.693366289138794, | |
| "learning_rate": 7.600000000000001e-06, | |
| "loss": 0.8611, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.2580645161290323, | |
| "grad_norm": 1.6432772874832153, | |
| "learning_rate": 7.800000000000002e-06, | |
| "loss": 0.8683, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.2903225806451613, | |
| "grad_norm": 1.7363226413726807, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.8663, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3225806451612903, | |
| "grad_norm": 1.8265684843063354, | |
| "learning_rate": 8.2e-06, | |
| "loss": 0.8635, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.3548387096774195, | |
| "grad_norm": 1.6863762140274048, | |
| "learning_rate": 8.400000000000001e-06, | |
| "loss": 0.8568, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.3870967741935485, | |
| "grad_norm": 1.7524291276931763, | |
| "learning_rate": 8.6e-06, | |
| "loss": 0.8877, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.4193548387096775, | |
| "grad_norm": 1.7783327102661133, | |
| "learning_rate": 8.8e-06, | |
| "loss": 0.8711, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.4516129032258065, | |
| "grad_norm": 1.8143388032913208, | |
| "learning_rate": 9e-06, | |
| "loss": 0.8735, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.4838709677419355, | |
| "grad_norm": 1.7480504512786865, | |
| "learning_rate": 9.200000000000002e-06, | |
| "loss": 0.8213, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.5161290322580645, | |
| "grad_norm": 1.7327862977981567, | |
| "learning_rate": 9.4e-06, | |
| "loss": 0.8095, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.5483870967741935, | |
| "grad_norm": 1.6738107204437256, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.8597, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.5483870967741935, | |
| "eval_loss": 1.065739393234253, | |
| "eval_runtime": 18.7181, | |
| "eval_samples_per_second": 41.404, | |
| "eval_steps_per_second": 5.182, | |
| "step": 48 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 62, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 16, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.748022277535957e+18, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |