| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9504950495049505, | |
| "eval_steps": 7, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.039603960396039604, | |
| "grad_norm": 0.2005040943622589, | |
| "learning_rate": 0.0001, | |
| "loss": 1.222, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.039603960396039604, | |
| "eval_loss": 1.2184828519821167, | |
| "eval_runtime": 135.46, | |
| "eval_samples_per_second": 7.412, | |
| "eval_steps_per_second": 0.93, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07920792079207921, | |
| "grad_norm": 0.19737568497657776, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2035, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.1188118811881188, | |
| "grad_norm": 0.22430424392223358, | |
| "learning_rate": 0.00019978589232386035, | |
| "loss": 1.2364, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.15841584158415842, | |
| "grad_norm": 0.3319801986217499, | |
| "learning_rate": 0.00019914448613738106, | |
| "loss": 1.1084, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.19801980198019803, | |
| "grad_norm": 0.2900150418281555, | |
| "learning_rate": 0.00019807852804032305, | |
| "loss": 1.0104, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.2376237623762376, | |
| "grad_norm": 0.3314291834831238, | |
| "learning_rate": 0.00019659258262890683, | |
| "loss": 0.9168, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.27722772277227725, | |
| "grad_norm": 0.4147993326187134, | |
| "learning_rate": 0.0001946930129495106, | |
| "loss": 0.9168, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.27722772277227725, | |
| "eval_loss": 0.8507946729660034, | |
| "eval_runtime": 135.7758, | |
| "eval_samples_per_second": 7.395, | |
| "eval_steps_per_second": 0.928, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.31683168316831684, | |
| "grad_norm": 0.32303664088249207, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.8745, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.3564356435643564, | |
| "grad_norm": 0.19321317970752716, | |
| "learning_rate": 0.00018968727415326884, | |
| "loss": 0.8174, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.39603960396039606, | |
| "grad_norm": 0.18172240257263184, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 0.8014, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.43564356435643564, | |
| "grad_norm": 0.13811971247196198, | |
| "learning_rate": 0.00018314696123025454, | |
| "loss": 0.7884, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4752475247524752, | |
| "grad_norm": 0.13254143297672272, | |
| "learning_rate": 0.00017933533402912354, | |
| "loss": 0.7617, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5148514851485149, | |
| "grad_norm": 0.10325208306312561, | |
| "learning_rate": 0.00017518398074789775, | |
| "loss": 0.7857, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5544554455445545, | |
| "grad_norm": 0.08808338642120361, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.7745, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5544554455445545, | |
| "eval_loss": 0.7478359937667847, | |
| "eval_runtime": 135.6094, | |
| "eval_samples_per_second": 7.404, | |
| "eval_steps_per_second": 0.929, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.594059405940594, | |
| "grad_norm": 0.0986776053905487, | |
| "learning_rate": 0.00016593458151000688, | |
| "loss": 0.7461, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6336633663366337, | |
| "grad_norm": 0.1371486485004425, | |
| "learning_rate": 0.00016087614290087208, | |
| "loss": 0.7296, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.6732673267326733, | |
| "grad_norm": 0.12637566030025482, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 0.776, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.7128712871287128, | |
| "grad_norm": 0.09393499791622162, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.7159, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7524752475247525, | |
| "grad_norm": 0.07980824261903763, | |
| "learning_rate": 0.00014422886902190014, | |
| "loss": 0.7219, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7920792079207921, | |
| "grad_norm": 0.07864726334810257, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.7282, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8316831683168316, | |
| "grad_norm": 0.08049978315830231, | |
| "learning_rate": 0.00013214394653031616, | |
| "loss": 0.7129, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8316831683168316, | |
| "eval_loss": 0.713334321975708, | |
| "eval_runtime": 135.7568, | |
| "eval_samples_per_second": 7.396, | |
| "eval_steps_per_second": 0.928, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8712871287128713, | |
| "grad_norm": 0.07534715533256531, | |
| "learning_rate": 0.00012588190451025207, | |
| "loss": 0.7116, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.9108910891089109, | |
| "grad_norm": 0.07699955999851227, | |
| "learning_rate": 0.00011950903220161285, | |
| "loss": 0.7159, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.9504950495049505, | |
| "grad_norm": 0.09450427442789078, | |
| "learning_rate": 0.00011305261922200519, | |
| "loss": 0.7165, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9900990099009901, | |
| "grad_norm": 0.08074381202459335, | |
| "learning_rate": 0.00010654031292301432, | |
| "loss": 0.6931, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0297029702970297, | |
| "grad_norm": 0.06008383259177208, | |
| "learning_rate": 0.0001, | |
| "loss": 0.689, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0396039603960396, | |
| "grad_norm": 0.05818495154380798, | |
| "learning_rate": 9.345968707698569e-05, | |
| "loss": 0.6787, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.0792079207920793, | |
| "grad_norm": 0.06621662527322769, | |
| "learning_rate": 8.694738077799488e-05, | |
| "loss": 0.7005, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.0792079207920793, | |
| "eval_loss": 0.6945069432258606, | |
| "eval_runtime": 135.5698, | |
| "eval_samples_per_second": 7.406, | |
| "eval_steps_per_second": 0.929, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.118811881188119, | |
| "grad_norm": 0.07372936606407166, | |
| "learning_rate": 8.049096779838719e-05, | |
| "loss": 0.6715, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.1584158415841583, | |
| "grad_norm": 0.07513756304979324, | |
| "learning_rate": 7.411809548974792e-05, | |
| "loss": 0.6807, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.198019801980198, | |
| "grad_norm": 0.06907230615615845, | |
| "learning_rate": 6.785605346968386e-05, | |
| "loss": 0.6782, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.2376237623762376, | |
| "grad_norm": 0.07009885460138321, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.6685, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.2772277227722773, | |
| "grad_norm": 0.06578447669744492, | |
| "learning_rate": 5.577113097809989e-05, | |
| "loss": 0.6894, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.316831683168317, | |
| "grad_norm": 0.0658915713429451, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.6888, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.3564356435643563, | |
| "grad_norm": 0.057827264070510864, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 0.6863, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.3564356435643563, | |
| "eval_loss": 0.6855179071426392, | |
| "eval_runtime": 135.8224, | |
| "eval_samples_per_second": 7.392, | |
| "eval_steps_per_second": 0.928, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.396039603960396, | |
| "grad_norm": 0.05925939604640007, | |
| "learning_rate": 3.9123857099127936e-05, | |
| "loss": 0.6888, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.4356435643564356, | |
| "grad_norm": 0.06030790135264397, | |
| "learning_rate": 3.406541848999312e-05, | |
| "loss": 0.6721, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.4752475247524752, | |
| "grad_norm": 0.06920236349105835, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.6801, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.5148514851485149, | |
| "grad_norm": 0.06107211485505104, | |
| "learning_rate": 2.4816019252102273e-05, | |
| "loss": 0.6484, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.5544554455445545, | |
| "grad_norm": 0.06593398004770279, | |
| "learning_rate": 2.0664665970876496e-05, | |
| "loss": 0.6821, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5940594059405941, | |
| "grad_norm": 0.058774009346961975, | |
| "learning_rate": 1.6853038769745467e-05, | |
| "loss": 0.6761, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6336633663366338, | |
| "grad_norm": 0.06204603239893913, | |
| "learning_rate": 1.339745962155613e-05, | |
| "loss": 0.6688, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6336633663366338, | |
| "eval_loss": 0.6808369159698486, | |
| "eval_runtime": 135.5705, | |
| "eval_samples_per_second": 7.406, | |
| "eval_steps_per_second": 0.929, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6732673267326734, | |
| "grad_norm": 0.057622041553258896, | |
| "learning_rate": 1.0312725846731175e-05, | |
| "loss": 0.6675, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.7128712871287128, | |
| "grad_norm": 0.05819413810968399, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.6867, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.7524752475247525, | |
| "grad_norm": 0.05817948281764984, | |
| "learning_rate": 5.306987050489442e-06, | |
| "loss": 0.6813, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.7920792079207921, | |
| "grad_norm": 0.058505669236183167, | |
| "learning_rate": 3.40741737109318e-06, | |
| "loss": 0.6702, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.8316831683168315, | |
| "grad_norm": 0.05659426376223564, | |
| "learning_rate": 1.921471959676957e-06, | |
| "loss": 0.6635, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.8712871287128712, | |
| "grad_norm": 0.06038310378789902, | |
| "learning_rate": 8.555138626189618e-07, | |
| "loss": 0.6519, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.9108910891089108, | |
| "grad_norm": 0.05925183743238449, | |
| "learning_rate": 2.141076761396521e-07, | |
| "loss": 0.6772, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.9108910891089108, | |
| "eval_loss": 0.6800007820129395, | |
| "eval_runtime": 135.613, | |
| "eval_samples_per_second": 7.403, | |
| "eval_steps_per_second": 0.929, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.9504950495049505, | |
| "grad_norm": 0.05675528571009636, | |
| "learning_rate": 0.0, | |
| "loss": 0.664, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 2.984041808658432e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |