| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.01054546413224012, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0001054546413224012, | |
| "eval_loss": 3.0851244926452637, | |
| "eval_runtime": 1296.9727, | |
| "eval_samples_per_second": 12.314, | |
| "eval_steps_per_second": 1.54, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0003163639239672036, | |
| "grad_norm": 3.364920139312744, | |
| "learning_rate": 3e-05, | |
| "loss": 3.1129, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0006327278479344072, | |
| "grad_norm": 3.629132032394409, | |
| "learning_rate": 6e-05, | |
| "loss": 2.9367, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0009490917719016108, | |
| "grad_norm": 2.8804471492767334, | |
| "learning_rate": 9e-05, | |
| "loss": 2.1674, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0009490917719016108, | |
| "eval_loss": 1.5537173748016357, | |
| "eval_runtime": 1301.5706, | |
| "eval_samples_per_second": 12.271, | |
| "eval_steps_per_second": 1.534, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0012654556958688144, | |
| "grad_norm": 3.045879602432251, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.2596, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.001581819619836018, | |
| "grad_norm": 2.454657793045044, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.9167, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0018981835438032215, | |
| "grad_norm": 2.024900197982788, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.5806, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0018981835438032215, | |
| "eval_loss": 0.6141843199729919, | |
| "eval_runtime": 1299.0748, | |
| "eval_samples_per_second": 12.294, | |
| "eval_steps_per_second": 1.537, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.002214547467770425, | |
| "grad_norm": 1.8922075033187866, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.487, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0025309113917376287, | |
| "grad_norm": 1.6137950420379639, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.473, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0028472753157048323, | |
| "grad_norm": 1.769119143486023, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.5101, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0028472753157048323, | |
| "eval_loss": 0.42539533972740173, | |
| "eval_runtime": 1301.7236, | |
| "eval_samples_per_second": 12.269, | |
| "eval_steps_per_second": 1.534, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.003163639239672036, | |
| "grad_norm": 1.2414625883102417, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.3457, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0034800031636392395, | |
| "grad_norm": 1.1694670915603638, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.3041, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.003796367087606443, | |
| "grad_norm": 1.5880285501480103, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.3631, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.003796367087606443, | |
| "eval_loss": 0.3508250117301941, | |
| "eval_runtime": 1301.98, | |
| "eval_samples_per_second": 12.267, | |
| "eval_steps_per_second": 1.534, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.004112731011573647, | |
| "grad_norm": 1.1332464218139648, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.3533, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.00442909493554085, | |
| "grad_norm": 1.8899226188659668, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.4094, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.004745458859508054, | |
| "grad_norm": 1.209465503692627, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.2916, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.004745458859508054, | |
| "eval_loss": 0.3128226399421692, | |
| "eval_runtime": 1301.9125, | |
| "eval_samples_per_second": 12.267, | |
| "eval_steps_per_second": 1.534, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0050618227834752574, | |
| "grad_norm": 1.3826940059661865, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.5031, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0053781867074424615, | |
| "grad_norm": 1.3125900030136108, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.2013, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.005694550631409665, | |
| "grad_norm": 1.3176207542419434, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.3433, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.005694550631409665, | |
| "eval_loss": 0.2877451777458191, | |
| "eval_runtime": 1302.1449, | |
| "eval_samples_per_second": 12.265, | |
| "eval_steps_per_second": 1.534, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.006010914555376869, | |
| "grad_norm": 1.3576031923294067, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.2684, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.006327278479344072, | |
| "grad_norm": 1.230389952659607, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.2711, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.006643642403311276, | |
| "grad_norm": 1.3148422241210938, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.382, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.006643642403311276, | |
| "eval_loss": 0.2793634235858917, | |
| "eval_runtime": 1301.6079, | |
| "eval_samples_per_second": 12.27, | |
| "eval_steps_per_second": 1.534, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.006960006327278479, | |
| "grad_norm": 1.0743589401245117, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.2639, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.007276370251245683, | |
| "grad_norm": 0.9537119269371033, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.2925, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.007592734175212886, | |
| "grad_norm": 1.0721756219863892, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.2189, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.007592734175212886, | |
| "eval_loss": 0.2621087431907654, | |
| "eval_runtime": 1299.3498, | |
| "eval_samples_per_second": 12.292, | |
| "eval_steps_per_second": 1.537, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.00790909809918009, | |
| "grad_norm": 1.1351699829101562, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.2786, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.008225462023147294, | |
| "grad_norm": 1.0633392333984375, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.2632, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.008541825947114497, | |
| "grad_norm": 1.0229682922363281, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.2519, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.008541825947114497, | |
| "eval_loss": 0.2528908848762512, | |
| "eval_runtime": 1301.3962, | |
| "eval_samples_per_second": 12.272, | |
| "eval_steps_per_second": 1.535, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.0088581898710817, | |
| "grad_norm": 0.9887948632240295, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.1914, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.009174553795048905, | |
| "grad_norm": 0.971612811088562, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.2243, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.009490917719016109, | |
| "grad_norm": 1.1088848114013672, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.3923, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.009490917719016109, | |
| "eval_loss": 0.2498694211244583, | |
| "eval_runtime": 1301.4191, | |
| "eval_samples_per_second": 12.272, | |
| "eval_steps_per_second": 1.534, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.009807281642983312, | |
| "grad_norm": 1.1288279294967651, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.1999, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.010123645566950515, | |
| "grad_norm": 1.1936079263687134, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.1988, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.01044000949091772, | |
| "grad_norm": 0.9933436512947083, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.4169, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.01044000949091772, | |
| "eval_loss": 0.2488851249217987, | |
| "eval_runtime": 1301.5736, | |
| "eval_samples_per_second": 12.271, | |
| "eval_steps_per_second": 1.534, | |
| "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": 7.811167717372723e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |