| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.05512679162072767, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0011025358324145535, | |
| "grad_norm": 23.39788055419922, | |
| "learning_rate": 2e-05, | |
| "loss": 7.1169, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0011025358324145535, | |
| "eval_loss": 1.7347309589385986, | |
| "eval_runtime": 22.9307, | |
| "eval_samples_per_second": 16.659, | |
| "eval_steps_per_second": 8.329, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002205071664829107, | |
| "grad_norm": 24.70276641845703, | |
| "learning_rate": 4e-05, | |
| "loss": 7.0268, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0033076074972436605, | |
| "grad_norm": 17.368816375732422, | |
| "learning_rate": 6e-05, | |
| "loss": 6.4277, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004410143329658214, | |
| "grad_norm": 13.83151626586914, | |
| "learning_rate": 8e-05, | |
| "loss": 6.8072, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.005512679162072767, | |
| "grad_norm": 16.472185134887695, | |
| "learning_rate": 0.0001, | |
| "loss": 6.4075, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.006615214994487321, | |
| "grad_norm": 13.921393394470215, | |
| "learning_rate": 0.00012, | |
| "loss": 5.3773, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.007717750826901874, | |
| "grad_norm": 14.343507766723633, | |
| "learning_rate": 0.00014, | |
| "loss": 5.888, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.008820286659316428, | |
| "grad_norm": 11.941823959350586, | |
| "learning_rate": 0.00016, | |
| "loss": 5.9528, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.009922822491730982, | |
| "grad_norm": 9.306382179260254, | |
| "learning_rate": 0.00018, | |
| "loss": 5.2994, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.011025358324145534, | |
| "grad_norm": 10.549939155578613, | |
| "learning_rate": 0.0002, | |
| "loss": 5.3721, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.012127894156560088, | |
| "grad_norm": 7.934646129608154, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 5.1655, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.013230429988974642, | |
| "grad_norm": 8.310248374938965, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 4.6376, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.014332965821389196, | |
| "grad_norm": 8.056012153625488, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 4.2708, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.014332965821389196, | |
| "eval_loss": 1.0936849117279053, | |
| "eval_runtime": 22.4087, | |
| "eval_samples_per_second": 17.047, | |
| "eval_steps_per_second": 8.523, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.015435501653803748, | |
| "grad_norm": 8.5448579788208, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 4.9839, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.016538037486218304, | |
| "grad_norm": 6.94588565826416, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 4.6092, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.017640573318632856, | |
| "grad_norm": 6.220713138580322, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 3.401, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.018743109151047408, | |
| "grad_norm": 6.417628288269043, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 4.0094, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.019845644983461964, | |
| "grad_norm": 7.87187385559082, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 3.9014, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.020948180815876516, | |
| "grad_norm": 5.41270112991333, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 2.7863, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.022050716648291068, | |
| "grad_norm": 7.535908222198486, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 3.3717, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.023153252480705624, | |
| "grad_norm": 6.384044647216797, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 3.7017, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.024255788313120176, | |
| "grad_norm": 7.927468299865723, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 4.0124, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.025358324145534728, | |
| "grad_norm": 7.174431800842285, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 3.9183, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.026460859977949284, | |
| "grad_norm": 4.970147609710693, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 2.5369, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.027563395810363836, | |
| "grad_norm": 5.759519100189209, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 3.3105, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.02866593164277839, | |
| "grad_norm": 5.858071327209473, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 2.8269, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.02866593164277839, | |
| "eval_loss": 0.7912609577178955, | |
| "eval_runtime": 22.3984, | |
| "eval_samples_per_second": 17.055, | |
| "eval_steps_per_second": 8.527, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.029768467475192944, | |
| "grad_norm": 4.645652770996094, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 3.259, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.030871003307607496, | |
| "grad_norm": 5.244138240814209, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 3.2757, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.03197353914002205, | |
| "grad_norm": 4.861637115478516, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 3.5816, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.03307607497243661, | |
| "grad_norm": 4.345144271850586, | |
| "learning_rate": 0.0001, | |
| "loss": 3.0694, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.034178610804851156, | |
| "grad_norm": 4.8563947677612305, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 2.4021, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.03528114663726571, | |
| "grad_norm": 4.445272445678711, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 2.8229, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.03638368246968027, | |
| "grad_norm": 6.865347862243652, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 2.493, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.037486218302094816, | |
| "grad_norm": 4.879561901092529, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 3.2766, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.03858875413450937, | |
| "grad_norm": 4.274408340454102, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 2.3649, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.03969128996692393, | |
| "grad_norm": 4.128203868865967, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 3.3395, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.040793825799338476, | |
| "grad_norm": 4.402691841125488, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 3.1077, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.04189636163175303, | |
| "grad_norm": 4.494605541229248, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 2.9641, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.04299889746416759, | |
| "grad_norm": 4.473241806030273, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 3.0826, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.04299889746416759, | |
| "eval_loss": 0.7266038656234741, | |
| "eval_runtime": 22.4254, | |
| "eval_samples_per_second": 17.034, | |
| "eval_steps_per_second": 8.517, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.044101433296582136, | |
| "grad_norm": 4.886397361755371, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 3.2276, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04520396912899669, | |
| "grad_norm": 4.3510332107543945, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 3.1627, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.04630650496141125, | |
| "grad_norm": 4.668325901031494, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 2.767, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.047409040793825796, | |
| "grad_norm": 4.192807197570801, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 2.6006, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.04851157662624035, | |
| "grad_norm": 3.6173996925354004, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 2.4846, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.04961411245865491, | |
| "grad_norm": 4.095741271972656, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 2.8095, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.050716648291069456, | |
| "grad_norm": 4.493954181671143, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 4.394, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.05181918412348401, | |
| "grad_norm": 4.2892584800720215, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 3.0646, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.05292171995589857, | |
| "grad_norm": 5.72744083404541, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 2.4661, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.05402425578831312, | |
| "grad_norm": 3.8518455028533936, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 3.37, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.05512679162072767, | |
| "grad_norm": 3.915316104888916, | |
| "learning_rate": 0.0, | |
| "loss": 2.9115, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.75267079258112e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |