| { |
| "best_global_step": 200, |
| "best_metric": 0.35487493872642517, |
| "best_model_checkpoint": "/workspace/anchor/olmo-compassion/train/checkpoint-200", |
| "epoch": 0.21407546160021407, |
| "eval_steps": 50, |
| "global_step": 200, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.005351886540005352, |
| "grad_norm": 7.84370756149292, |
| "learning_rate": 1.7021276595744682e-05, |
| "loss": 7.033, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.010703773080010704, |
| "grad_norm": 6.463988304138184, |
| "learning_rate": 3.829787234042553e-05, |
| "loss": 6.5074, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.016055659620016056, |
| "grad_norm": 3.8663952350616455, |
| "learning_rate": 5.9574468085106384e-05, |
| "loss": 5.5202, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.02140754616002141, |
| "grad_norm": 3.429267168045044, |
| "learning_rate": 8.085106382978723e-05, |
| "loss": 4.6312, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.02675943270002676, |
| "grad_norm": 2.4048895835876465, |
| "learning_rate": 0.00010212765957446809, |
| "loss": 3.9047, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.03211131924003211, |
| "grad_norm": 1.6516971588134766, |
| "learning_rate": 0.00012340425531914893, |
| "loss": 3.239, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.037463205780037465, |
| "grad_norm": 1.9744592905044556, |
| "learning_rate": 0.0001446808510638298, |
| "loss": 2.8394, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.04281509232004282, |
| "grad_norm": 1.9995774030685425, |
| "learning_rate": 0.00016595744680851065, |
| "loss": 1.9605, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.048166978860048164, |
| "grad_norm": 1.9445796012878418, |
| "learning_rate": 0.0001872340425531915, |
| "loss": 1.4049, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.05351886540005352, |
| "grad_norm": 1.7529069185256958, |
| "learning_rate": 0.00019999749676283775, |
| "loss": 1.1704, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.05351886540005352, |
| "eval_loss": 0.7856047749519348, |
| "eval_runtime": 169.4008, |
| "eval_samples_per_second": 2.267, |
| "eval_steps_per_second": 1.133, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.05887075194005887, |
| "grad_norm": 1.2156940698623657, |
| "learning_rate": 0.00019996933678400946, |
| "loss": 0.8596, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.06422263848006422, |
| "grad_norm": 1.3200056552886963, |
| "learning_rate": 0.00019990989662046818, |
| "loss": 0.8833, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.06957452502006957, |
| "grad_norm": 1.5341733694076538, |
| "learning_rate": 0.00019981919487090972, |
| "loss": 0.8131, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.07492641156007493, |
| "grad_norm": 1.1513314247131348, |
| "learning_rate": 0.00019969725991571128, |
| "loss": 0.6417, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.08027829810008028, |
| "grad_norm": 0.8719304800033569, |
| "learning_rate": 0.00019954412990805107, |
| "loss": 0.8695, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.08563018464008564, |
| "grad_norm": 0.7839741706848145, |
| "learning_rate": 0.0001993598527619703, |
| "loss": 0.6572, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.09098207118009098, |
| "grad_norm": 0.7421998381614685, |
| "learning_rate": 0.00019914448613738106, |
| "loss": 0.782, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.09633395772009633, |
| "grad_norm": 0.747768759727478, |
| "learning_rate": 0.00019889809742202455, |
| "loss": 0.7823, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.10168584426010169, |
| "grad_norm": 0.7919906973838806, |
| "learning_rate": 0.0001986207637103857, |
| "loss": 0.6102, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.10703773080010703, |
| "grad_norm": 0.6297110915184021, |
| "learning_rate": 0.00019831257177957044, |
| "loss": 0.7222, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.10703773080010703, |
| "eval_loss": 0.42398348450660706, |
| "eval_runtime": 64.9622, |
| "eval_samples_per_second": 5.911, |
| "eval_steps_per_second": 2.956, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.1123896173401124, |
| "grad_norm": 0.6574537754058838, |
| "learning_rate": 0.00019797361806215332, |
| "loss": 0.6892, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.11774150388011774, |
| "grad_norm": 0.7165386080741882, |
| "learning_rate": 0.000197604008616004, |
| "loss": 0.6953, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.1230933904201231, |
| "grad_norm": 0.7242662310600281, |
| "learning_rate": 0.00019720385909110198, |
| "loss": 0.7986, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.12844527696012845, |
| "grad_norm": 0.6067697405815125, |
| "learning_rate": 0.0001967732946933499, |
| "loss": 0.7409, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.1337971635001338, |
| "grad_norm": 0.723360002040863, |
| "learning_rate": 0.00019631245014539698, |
| "loss": 0.4524, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.13914905004013914, |
| "grad_norm": 0.7545053362846375, |
| "learning_rate": 0.0001958214696444846, |
| "loss": 0.5135, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.1445009365801445, |
| "grad_norm": 0.5811852812767029, |
| "learning_rate": 0.0001953005068173272, |
| "loss": 0.4958, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.14985282312014986, |
| "grad_norm": 0.7165131568908691, |
| "learning_rate": 0.00019474972467204297, |
| "loss": 0.5611, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.1552047096601552, |
| "grad_norm": 0.8769603371620178, |
| "learning_rate": 0.00019416929554714888, |
| "loss": 0.8686, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.16055659620016055, |
| "grad_norm": 0.6099892258644104, |
| "learning_rate": 0.0001935594010576362, |
| "loss": 0.5971, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.16055659620016055, |
| "eval_loss": 0.37071776390075684, |
| "eval_runtime": 136.3832, |
| "eval_samples_per_second": 2.816, |
| "eval_steps_per_second": 1.408, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.1659084827401659, |
| "grad_norm": 0.6321601867675781, |
| "learning_rate": 0.000192920232038144, |
| "loss": 0.5999, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.17126036928017127, |
| "grad_norm": 0.6136798858642578, |
| "learning_rate": 0.0001922519884832469, |
| "loss": 0.6562, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.17661225582017662, |
| "grad_norm": 0.6746335625648499, |
| "learning_rate": 0.00019155487948487748, |
| "loss": 0.533, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.18196414236018196, |
| "grad_norm": 0.766614556312561, |
| "learning_rate": 0.0001908291231669019, |
| "loss": 0.7409, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.1873160289001873, |
| "grad_norm": 0.5589352250099182, |
| "learning_rate": 0.00019007494661686935, |
| "loss": 0.4672, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.19266791544019266, |
| "grad_norm": 0.59783935546875, |
| "learning_rate": 0.00018929258581495685, |
| "loss": 0.5064, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.19801980198019803, |
| "grad_norm": 0.6479551792144775, |
| "learning_rate": 0.0001884822855601316, |
| "loss": 0.6937, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.20337168852020338, |
| "grad_norm": 0.6843699812889099, |
| "learning_rate": 0.00018764429939355392, |
| "loss": 0.5394, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.20872357506020872, |
| "grad_norm": 0.5675429105758667, |
| "learning_rate": 0.00018677888951924474, |
| "loss": 0.7568, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.21407546160021407, |
| "grad_norm": 0.659488320350647, |
| "learning_rate": 0.00018588632672204264, |
| "loss": 0.4871, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.21407546160021407, |
| "eval_loss": 0.35487493872642517, |
| "eval_runtime": 65.702, |
| "eval_samples_per_second": 5.845, |
| "eval_steps_per_second": 2.922, |
| "step": 200 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 935, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 100, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 2, |
| "early_stopping_threshold": 0.001 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 0 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 5.501220870594724e+17, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
