| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.37453183520599254, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.003745318352059925, | |
| "eval_loss": 2.6081457138061523, | |
| "eval_runtime": 47.6507, | |
| "eval_samples_per_second": 9.444, | |
| "eval_steps_per_second": 1.196, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011235955056179775, | |
| "grad_norm": 2.9735543727874756, | |
| "learning_rate": 1.5e-05, | |
| "loss": 10.1416, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.02247191011235955, | |
| "grad_norm": 3.5065205097198486, | |
| "learning_rate": 3e-05, | |
| "loss": 10.634, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.033707865168539325, | |
| "grad_norm": 3.2851760387420654, | |
| "learning_rate": 4.5e-05, | |
| "loss": 10.0473, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.033707865168539325, | |
| "eval_loss": 2.5230321884155273, | |
| "eval_runtime": 48.9294, | |
| "eval_samples_per_second": 9.197, | |
| "eval_steps_per_second": 1.165, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0449438202247191, | |
| "grad_norm": 2.6816937923431396, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 9.8213, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.056179775280898875, | |
| "grad_norm": 2.483027458190918, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 9.4865, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.06741573033707865, | |
| "grad_norm": 2.5238866806030273, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 8.9648, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.06741573033707865, | |
| "eval_loss": 2.1459028720855713, | |
| "eval_runtime": 48.9218, | |
| "eval_samples_per_second": 9.198, | |
| "eval_steps_per_second": 1.165, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07865168539325842, | |
| "grad_norm": 2.4473624229431152, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 8.6589, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0898876404494382, | |
| "grad_norm": 2.805630683898926, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 8.0759, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.10112359550561797, | |
| "grad_norm": 2.6223995685577393, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 8.0252, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.10112359550561797, | |
| "eval_loss": 1.8734335899353027, | |
| "eval_runtime": 48.93, | |
| "eval_samples_per_second": 9.197, | |
| "eval_steps_per_second": 1.165, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.11235955056179775, | |
| "grad_norm": 2.5903525352478027, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 7.4746, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12359550561797752, | |
| "grad_norm": 2.3330588340759277, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 7.2413, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.1348314606741573, | |
| "grad_norm": 2.1777396202087402, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 6.7145, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.1348314606741573, | |
| "eval_loss": 1.7051557302474976, | |
| "eval_runtime": 48.9544, | |
| "eval_samples_per_second": 9.192, | |
| "eval_steps_per_second": 1.164, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.14606741573033707, | |
| "grad_norm": 2.583069324493408, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 6.6727, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.15730337078651685, | |
| "grad_norm": 2.388641119003296, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 6.2881, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.16853932584269662, | |
| "grad_norm": 2.6823623180389404, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 7.006, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.16853932584269662, | |
| "eval_loss": 1.584166169166565, | |
| "eval_runtime": 48.8986, | |
| "eval_samples_per_second": 9.203, | |
| "eval_steps_per_second": 1.166, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.1797752808988764, | |
| "grad_norm": 3.2928547859191895, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 6.5587, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.19101123595505617, | |
| "grad_norm": 2.7404942512512207, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 6.3989, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.20224719101123595, | |
| "grad_norm": 2.7708077430725098, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 6.1356, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.20224719101123595, | |
| "eval_loss": 1.492134690284729, | |
| "eval_runtime": 48.924, | |
| "eval_samples_per_second": 9.198, | |
| "eval_steps_per_second": 1.165, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.21348314606741572, | |
| "grad_norm": 2.9620726108551025, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 5.8867, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.2247191011235955, | |
| "grad_norm": 3.2739944458007812, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 5.3803, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.23595505617977527, | |
| "grad_norm": 3.1160500049591064, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 5.8344, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.23595505617977527, | |
| "eval_loss": 1.4242901802062988, | |
| "eval_runtime": 48.9559, | |
| "eval_samples_per_second": 9.192, | |
| "eval_steps_per_second": 1.164, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.24719101123595505, | |
| "grad_norm": 3.311922311782837, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 5.9825, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.25842696629213485, | |
| "grad_norm": 3.3331925868988037, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 5.403, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.2696629213483146, | |
| "grad_norm": 3.360844373703003, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 5.582, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.2696629213483146, | |
| "eval_loss": 1.383041262626648, | |
| "eval_runtime": 48.9452, | |
| "eval_samples_per_second": 9.194, | |
| "eval_steps_per_second": 1.165, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.2808988764044944, | |
| "grad_norm": 3.310269355773926, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 5.4252, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.29213483146067415, | |
| "grad_norm": 3.2484939098358154, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 5.3267, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.30337078651685395, | |
| "grad_norm": 3.0837814807891846, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 5.3015, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.30337078651685395, | |
| "eval_loss": 1.3595863580703735, | |
| "eval_runtime": 48.9241, | |
| "eval_samples_per_second": 9.198, | |
| "eval_steps_per_second": 1.165, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.3146067415730337, | |
| "grad_norm": 3.24570631980896, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 5.4681, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.3258426966292135, | |
| "grad_norm": 3.1700186729431152, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 5.1478, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.33707865168539325, | |
| "grad_norm": 3.2288196086883545, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 5.9653, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.33707865168539325, | |
| "eval_loss": 1.3506044149398804, | |
| "eval_runtime": 48.915, | |
| "eval_samples_per_second": 9.2, | |
| "eval_steps_per_second": 1.165, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.34831460674157305, | |
| "grad_norm": 3.245680093765259, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 6.1023, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.3595505617977528, | |
| "grad_norm": 2.9969637393951416, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 5.7552, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.3707865168539326, | |
| "grad_norm": 3.160747766494751, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 5.9094, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.3707865168539326, | |
| "eval_loss": 1.3485381603240967, | |
| "eval_runtime": 48.9177, | |
| "eval_samples_per_second": 9.199, | |
| "eval_steps_per_second": 1.165, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.950140184756224e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |