| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.07414272474513438, | |
| "eval_steps": 25, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0014828544949026877, | |
| "grad_norm": 1.846521019935608, | |
| "learning_rate": 5e-05, | |
| "loss": 5.445, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0014828544949026877, | |
| "eval_loss": 5.937737941741943, | |
| "eval_runtime": 3.4489, | |
| "eval_samples_per_second": 14.497, | |
| "eval_steps_per_second": 3.769, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0029657089898053753, | |
| "grad_norm": 1.7791767120361328, | |
| "learning_rate": 0.0001, | |
| "loss": 5.5437, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.004448563484708063, | |
| "grad_norm": 0.9079276919364929, | |
| "learning_rate": 9.990365154573717e-05, | |
| "loss": 5.4807, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.005931417979610751, | |
| "grad_norm": 1.3516089916229248, | |
| "learning_rate": 9.961501876182148e-05, | |
| "loss": 5.43, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0074142724745134385, | |
| "grad_norm": 0.9882570505142212, | |
| "learning_rate": 9.913533761814537e-05, | |
| "loss": 5.414, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.008897126969416125, | |
| "grad_norm": 0.5981724262237549, | |
| "learning_rate": 9.846666218300807e-05, | |
| "loss": 5.4133, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.010379981464318814, | |
| "grad_norm": 0.7474287748336792, | |
| "learning_rate": 9.761185582727977e-05, | |
| "loss": 5.4478, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.011862835959221501, | |
| "grad_norm": 0.9182415008544922, | |
| "learning_rate": 9.657457896300791e-05, | |
| "loss": 5.4144, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.013345690454124188, | |
| "grad_norm": 1.5489169359207153, | |
| "learning_rate": 9.535927336897098e-05, | |
| "loss": 5.4635, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.014828544949026877, | |
| "grad_norm": 1.1340339183807373, | |
| "learning_rate": 9.397114317029975e-05, | |
| "loss": 5.4747, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.016311399443929564, | |
| "grad_norm": 1.5122523307800293, | |
| "learning_rate": 9.241613255361455e-05, | |
| "loss": 5.4525, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.01779425393883225, | |
| "grad_norm": 5.111401557922363, | |
| "learning_rate": 9.070090031310558e-05, | |
| "loss": 5.5327, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.01927710843373494, | |
| "grad_norm": 9.320221900939941, | |
| "learning_rate": 8.883279133655399e-05, | |
| "loss": 5.7383, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.02075996292863763, | |
| "grad_norm": 3.333993434906006, | |
| "learning_rate": 8.681980515339464e-05, | |
| "loss": 5.335, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.022242817423540315, | |
| "grad_norm": 1.4352670907974243, | |
| "learning_rate": 8.467056167950311e-05, | |
| "loss": 5.3649, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.023725671918443002, | |
| "grad_norm": 1.439532995223999, | |
| "learning_rate": 8.239426430539243e-05, | |
| "loss": 5.3677, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.02520852641334569, | |
| "grad_norm": 0.853127121925354, | |
| "learning_rate": 8.000066048588211e-05, | |
| "loss": 5.3222, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.026691380908248376, | |
| "grad_norm": 0.7488646507263184, | |
| "learning_rate": 7.75e-05, | |
| "loss": 5.3506, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.028174235403151067, | |
| "grad_norm": 0.9415917992591858, | |
| "learning_rate": 7.490299105985507e-05, | |
| "loss": 5.3019, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.029657089898053754, | |
| "grad_norm": 0.8569623827934265, | |
| "learning_rate": 7.222075445642904e-05, | |
| "loss": 5.3242, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03113994439295644, | |
| "grad_norm": 0.8924872279167175, | |
| "learning_rate": 6.946477593864228e-05, | |
| "loss": 5.2043, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.03262279888785913, | |
| "grad_norm": 1.061047911643982, | |
| "learning_rate": 6.664685702961344e-05, | |
| "loss": 5.2871, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.034105653382761815, | |
| "grad_norm": 1.4060221910476685, | |
| "learning_rate": 6.377906449072578e-05, | |
| "loss": 5.3072, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0355885078776645, | |
| "grad_norm": 1.084276556968689, | |
| "learning_rate": 6.087367864990233e-05, | |
| "loss": 5.2147, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03707136237256719, | |
| "grad_norm": 1.1851881742477417, | |
| "learning_rate": 5.794314081535644e-05, | |
| "loss": 5.2402, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03707136237256719, | |
| "eval_loss": 5.7982354164123535, | |
| "eval_runtime": 3.4791, | |
| "eval_samples_per_second": 14.372, | |
| "eval_steps_per_second": 3.737, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03855421686746988, | |
| "grad_norm": 15.813837051391602, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 5.8473, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.04003707136237257, | |
| "grad_norm": 13.459531784057617, | |
| "learning_rate": 5.205685918464356e-05, | |
| "loss": 5.7006, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.04151992585727526, | |
| "grad_norm": 7.003750324249268, | |
| "learning_rate": 4.912632135009769e-05, | |
| "loss": 5.5551, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.043002780352177944, | |
| "grad_norm": 2.730677843093872, | |
| "learning_rate": 4.6220935509274235e-05, | |
| "loss": 5.4208, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.04448563484708063, | |
| "grad_norm": 1.527279019355774, | |
| "learning_rate": 4.3353142970386564e-05, | |
| "loss": 5.3499, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04596848934198332, | |
| "grad_norm": 1.468549132347107, | |
| "learning_rate": 4.053522406135775e-05, | |
| "loss": 5.3628, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.047451343836886005, | |
| "grad_norm": 1.2678706645965576, | |
| "learning_rate": 3.777924554357096e-05, | |
| "loss": 5.3599, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.04893419833178869, | |
| "grad_norm": 1.247628092765808, | |
| "learning_rate": 3.509700894014496e-05, | |
| "loss": 5.3066, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.05041705282669138, | |
| "grad_norm": 1.016780972480774, | |
| "learning_rate": 3.250000000000001e-05, | |
| "loss": 5.3093, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.051899907321594066, | |
| "grad_norm": 1.0291693210601807, | |
| "learning_rate": 2.9999339514117912e-05, | |
| "loss": 5.32, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.05338276181649675, | |
| "grad_norm": 1.1711857318878174, | |
| "learning_rate": 2.760573569460757e-05, | |
| "loss": 5.3058, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.05486561631139945, | |
| "grad_norm": 1.3060027360916138, | |
| "learning_rate": 2.53294383204969e-05, | |
| "loss": 5.3427, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.056348470806302134, | |
| "grad_norm": 1.1797828674316406, | |
| "learning_rate": 2.3180194846605367e-05, | |
| "loss": 5.2493, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.05783132530120482, | |
| "grad_norm": 1.2685225009918213, | |
| "learning_rate": 2.1167208663446025e-05, | |
| "loss": 5.2396, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.05931417979610751, | |
| "grad_norm": 1.1514685153961182, | |
| "learning_rate": 1.9299099686894423e-05, | |
| "loss": 5.1973, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.060797034291010195, | |
| "grad_norm": 0.915943443775177, | |
| "learning_rate": 1.758386744638546e-05, | |
| "loss": 5.2604, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.06227988878591288, | |
| "grad_norm": 0.7309096455574036, | |
| "learning_rate": 1.602885682970026e-05, | |
| "loss": 5.2574, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.06376274328081558, | |
| "grad_norm": 0.6121890544891357, | |
| "learning_rate": 1.464072663102903e-05, | |
| "loss": 5.237, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.06524559777571826, | |
| "grad_norm": 0.5316140651702881, | |
| "learning_rate": 1.3425421036992098e-05, | |
| "loss": 5.2674, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.06672845227062095, | |
| "grad_norm": 0.5401632189750671, | |
| "learning_rate": 1.2388144172720251e-05, | |
| "loss": 5.2222, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06821130676552363, | |
| "grad_norm": 1.0013433694839478, | |
| "learning_rate": 1.1533337816991932e-05, | |
| "loss": 5.1911, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.06969416126042632, | |
| "grad_norm": 0.9075887203216553, | |
| "learning_rate": 1.0864662381854632e-05, | |
| "loss": 5.2855, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.071177015755329, | |
| "grad_norm": 1.1342015266418457, | |
| "learning_rate": 1.0384981238178534e-05, | |
| "loss": 5.2796, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0726598702502317, | |
| "grad_norm": 1.9507712125778198, | |
| "learning_rate": 1.0096348454262845e-05, | |
| "loss": 5.2357, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.07414272474513438, | |
| "grad_norm": 2.180121898651123, | |
| "learning_rate": 1e-05, | |
| "loss": 5.3245, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07414272474513438, | |
| "eval_loss": 5.244147777557373, | |
| "eval_runtime": 3.4601, | |
| "eval_samples_per_second": 14.45, | |
| "eval_steps_per_second": 3.757, | |
| "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": 5.968083617316864e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |