| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.003926303287297427, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.9263032872974274e-05, | |
| "eval_loss": 3.7315213680267334, | |
| "eval_runtime": 773.5499, | |
| "eval_samples_per_second": 27.727, | |
| "eval_steps_per_second": 6.932, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00011778909861892282, | |
| "grad_norm": 31.5966854095459, | |
| "learning_rate": 3e-05, | |
| "loss": 3.5372, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00023557819723784564, | |
| "grad_norm": 11.457754135131836, | |
| "learning_rate": 6e-05, | |
| "loss": 1.4271, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00035336729585676844, | |
| "grad_norm": 0.03152690827846527, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0468, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00035336729585676844, | |
| "eval_loss": 0.0013787680072709918, | |
| "eval_runtime": 777.5058, | |
| "eval_samples_per_second": 27.586, | |
| "eval_steps_per_second": 6.896, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0004711563944756913, | |
| "grad_norm": 0.14033684134483337, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.0189, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0005889454930946141, | |
| "grad_norm": 0.008016275241971016, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.0107, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0007067345917135369, | |
| "grad_norm": 0.9549577236175537, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.0004, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0007067345917135369, | |
| "eval_loss": 0.0004475166497286409, | |
| "eval_runtime": 777.492, | |
| "eval_samples_per_second": 27.586, | |
| "eval_steps_per_second": 6.897, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0008245236903324598, | |
| "grad_norm": 0.030346207320690155, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.0001, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0009423127889513826, | |
| "grad_norm": 0.0005559443379752338, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0010601018875703054, | |
| "grad_norm": 6.331143231363967e-05, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0010601018875703054, | |
| "eval_loss": 0.000494947365950793, | |
| "eval_runtime": 777.5031, | |
| "eval_samples_per_second": 27.586, | |
| "eval_steps_per_second": 6.896, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0011778909861892282, | |
| "grad_norm": 4.403312050271779e-05, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.001295680084808151, | |
| "grad_norm": 3.843701051664539e-05, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0014134691834270738, | |
| "grad_norm": 4.2941730498569086e-05, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0014134691834270738, | |
| "eval_loss": 0.0005166567279957235, | |
| "eval_runtime": 777.5045, | |
| "eval_samples_per_second": 27.586, | |
| "eval_steps_per_second": 6.896, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0015312582820459966, | |
| "grad_norm": 3.606522659538314e-05, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0016490473806649196, | |
| "grad_norm": 3.24067659676075e-05, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0017668364792838424, | |
| "grad_norm": 2.977346593979746e-05, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0017668364792838424, | |
| "eval_loss": 0.0005258838646113873, | |
| "eval_runtime": 777.4979, | |
| "eval_samples_per_second": 27.586, | |
| "eval_steps_per_second": 6.896, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0018846255779027652, | |
| "grad_norm": 3.105065115960315e-05, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0020024146765216877, | |
| "grad_norm": 3.0599792808061466e-05, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.0, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.0021202037751406107, | |
| "grad_norm": 3.959710738854483e-05, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.0, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.0021202037751406107, | |
| "eval_loss": 0.0005288115935400128, | |
| "eval_runtime": 777.4935, | |
| "eval_samples_per_second": 27.586, | |
| "eval_steps_per_second": 6.897, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.0022379928737595338, | |
| "grad_norm": 3.3493728551547974e-05, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.0, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.0023557819723784563, | |
| "grad_norm": 3.4933964343508705e-05, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0024735710709973793, | |
| "grad_norm": 2.990081520692911e-05, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.0, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0024735710709973793, | |
| "eval_loss": 0.0005307694664224982, | |
| "eval_runtime": 777.489, | |
| "eval_samples_per_second": 27.586, | |
| "eval_steps_per_second": 6.897, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.002591360169616302, | |
| "grad_norm": 3.213492163922638e-05, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.0, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.002709149268235225, | |
| "grad_norm": 2.8474241844378412e-05, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.0, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.0028269383668541475, | |
| "grad_norm": 3.170351919834502e-05, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.0, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0028269383668541475, | |
| "eval_loss": 0.0005312738940119743, | |
| "eval_runtime": 777.5031, | |
| "eval_samples_per_second": 27.586, | |
| "eval_steps_per_second": 6.896, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0029447274654730705, | |
| "grad_norm": 3.209977876394987e-05, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.003062516564091993, | |
| "grad_norm": 2.9874341635149904e-05, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.0, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.003180305662710916, | |
| "grad_norm": 3.113800266874023e-05, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.0, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.003180305662710916, | |
| "eval_loss": 0.0005310006672516465, | |
| "eval_runtime": 777.5014, | |
| "eval_samples_per_second": 27.586, | |
| "eval_steps_per_second": 6.896, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.003298094761329839, | |
| "grad_norm": 3.5468474379740655e-05, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.0, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.0034158838599487617, | |
| "grad_norm": 2.9161385100451298e-05, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.0, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.0035336729585676847, | |
| "grad_norm": 2.8734202714986168e-05, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0035336729585676847, | |
| "eval_loss": 0.0005303475772961974, | |
| "eval_runtime": 777.4997, | |
| "eval_samples_per_second": 27.586, | |
| "eval_steps_per_second": 6.896, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0036514620571866073, | |
| "grad_norm": 2.96364141831873e-05, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.0, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.0037692511558055303, | |
| "grad_norm": 3.082916373386979e-05, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.0, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.003887040254424453, | |
| "grad_norm": 2.8683465643553063e-05, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.0, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.003887040254424453, | |
| "eval_loss": 0.0005314745358191431, | |
| "eval_runtime": 777.4915, | |
| "eval_samples_per_second": 27.586, | |
| "eval_steps_per_second": 6.897, | |
| "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": 2.02538779410432e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |