| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5710206995003569, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005710206995003569, | |
| "eval_loss": 4.834342956542969, | |
| "eval_runtime": 65.5166, | |
| "eval_samples_per_second": 9.005, | |
| "eval_steps_per_second": 4.503, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.017130620985010708, | |
| "grad_norm": 3.7000343799591064, | |
| "learning_rate": 3e-05, | |
| "loss": 4.5765, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.034261241970021415, | |
| "grad_norm": 4.092134475708008, | |
| "learning_rate": 6e-05, | |
| "loss": 4.5819, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.05139186295503212, | |
| "grad_norm": 4.719603538513184, | |
| "learning_rate": 9e-05, | |
| "loss": 3.6777, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05139186295503212, | |
| "eval_loss": 3.054090738296509, | |
| "eval_runtime": 65.7842, | |
| "eval_samples_per_second": 8.969, | |
| "eval_steps_per_second": 4.484, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.06852248394004283, | |
| "grad_norm": 3.8146257400512695, | |
| "learning_rate": 0.00012, | |
| "loss": 2.8501, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.08565310492505353, | |
| "grad_norm": 2.3158440589904785, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 2.4364, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.10278372591006424, | |
| "grad_norm": 1.6497728824615479, | |
| "learning_rate": 0.00018, | |
| "loss": 2.1518, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.10278372591006424, | |
| "eval_loss": 2.1508984565734863, | |
| "eval_runtime": 65.7679, | |
| "eval_samples_per_second": 8.971, | |
| "eval_steps_per_second": 4.485, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.11991434689507495, | |
| "grad_norm": 1.816499948501587, | |
| "learning_rate": 0.0001999229036240723, | |
| "loss": 2.1273, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.13704496788008566, | |
| "grad_norm": 2.241565227508545, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 2.0534, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.15417558886509636, | |
| "grad_norm": 1.3755393028259277, | |
| "learning_rate": 0.00019624552364536473, | |
| "loss": 2.0302, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.15417558886509636, | |
| "eval_loss": 2.0600290298461914, | |
| "eval_runtime": 65.7463, | |
| "eval_samples_per_second": 8.974, | |
| "eval_steps_per_second": 4.487, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.17130620985010706, | |
| "grad_norm": 1.3961236476898193, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 2.0442, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.18843683083511778, | |
| "grad_norm": 1.2159768342971802, | |
| "learning_rate": 0.00018724960070727972, | |
| "loss": 1.9398, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.20556745182012848, | |
| "grad_norm": 1.243770718574524, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 2.0249, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.20556745182012848, | |
| "eval_loss": 2.0228545665740967, | |
| "eval_runtime": 65.7648, | |
| "eval_samples_per_second": 8.971, | |
| "eval_steps_per_second": 4.486, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.22269807280513917, | |
| "grad_norm": 1.321190595626831, | |
| "learning_rate": 0.00017343225094356855, | |
| "loss": 2.0458, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.2398286937901499, | |
| "grad_norm": 1.0691664218902588, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 1.9291, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.2569593147751606, | |
| "grad_norm": 1.1871341466903687, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 1.9359, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2569593147751606, | |
| "eval_loss": 2.0000336170196533, | |
| "eval_runtime": 65.6889, | |
| "eval_samples_per_second": 8.982, | |
| "eval_steps_per_second": 4.491, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2740899357601713, | |
| "grad_norm": 1.500360131263733, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 2.0105, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.291220556745182, | |
| "grad_norm": 1.301650881767273, | |
| "learning_rate": 0.0001346117057077493, | |
| "loss": 1.9484, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.3083511777301927, | |
| "grad_norm": 1.1405830383300781, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 1.8932, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.3083511777301927, | |
| "eval_loss": 1.986473798751831, | |
| "eval_runtime": 65.7166, | |
| "eval_samples_per_second": 8.978, | |
| "eval_steps_per_second": 4.489, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.32548179871520344, | |
| "grad_norm": 1.0970661640167236, | |
| "learning_rate": 0.00011175373974578378, | |
| "loss": 1.9572, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.3426124197002141, | |
| "grad_norm": 1.1758753061294556, | |
| "learning_rate": 0.0001, | |
| "loss": 2.0218, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.35974304068522484, | |
| "grad_norm": 1.2941383123397827, | |
| "learning_rate": 8.824626025421626e-05, | |
| "loss": 1.9693, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.35974304068522484, | |
| "eval_loss": 1.9790188074111938, | |
| "eval_runtime": 65.5469, | |
| "eval_samples_per_second": 9.001, | |
| "eval_steps_per_second": 4.501, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.37687366167023556, | |
| "grad_norm": 1.0595579147338867, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 1.9072, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.39400428265524623, | |
| "grad_norm": 1.2764127254486084, | |
| "learning_rate": 6.538829429225069e-05, | |
| "loss": 2.0075, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.41113490364025695, | |
| "grad_norm": 1.0607476234436035, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 1.9258, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.41113490364025695, | |
| "eval_loss": 1.9672698974609375, | |
| "eval_runtime": 65.5425, | |
| "eval_samples_per_second": 9.002, | |
| "eval_steps_per_second": 4.501, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.4282655246252677, | |
| "grad_norm": 1.078697919845581, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 1.9402, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.44539614561027835, | |
| "grad_norm": 1.0098077058792114, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 1.947, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.4625267665952891, | |
| "grad_norm": 1.063792109489441, | |
| "learning_rate": 2.6567749056431467e-05, | |
| "loss": 1.9917, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.4625267665952891, | |
| "eval_loss": 1.962007999420166, | |
| "eval_runtime": 65.5561, | |
| "eval_samples_per_second": 9.0, | |
| "eval_steps_per_second": 4.5, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.4796573875802998, | |
| "grad_norm": 1.0382856130599976, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 2.0827, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.49678800856531047, | |
| "grad_norm": 1.0031726360321045, | |
| "learning_rate": 1.2750399292720283e-05, | |
| "loss": 1.9068, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.5139186295503212, | |
| "grad_norm": 1.1354076862335205, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 1.9417, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5139186295503212, | |
| "eval_loss": 1.9562921524047852, | |
| "eval_runtime": 65.5479, | |
| "eval_samples_per_second": 9.001, | |
| "eval_steps_per_second": 4.501, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5310492505353319, | |
| "grad_norm": 1.0723165273666382, | |
| "learning_rate": 3.7544763546352834e-06, | |
| "loss": 1.894, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.5481798715203426, | |
| "grad_norm": 1.0611053705215454, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 1.8989, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.5653104925053534, | |
| "grad_norm": 1.0602564811706543, | |
| "learning_rate": 7.709637592770991e-08, | |
| "loss": 1.939, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.5653104925053534, | |
| "eval_loss": 1.9549332857131958, | |
| "eval_runtime": 65.5189, | |
| "eval_samples_per_second": 9.005, | |
| "eval_steps_per_second": 4.503, | |
| "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": 5.968083617316864e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |